date stringlengths 10 10 | nb_tokens int64 60 629k | text_size int64 234 1.02M | content stringlengths 234 1.02M |
|---|---|---|---|
2019/01/14 | 614 | 2,465 | <issue_start>username_0: I was wondering if machine learning algorithms (CNNs?) can be used/trained to differentiate between small differences in details between images (such as slight differences in shades of red or other colours, or the presence of small objects between otherwise very similar images?)? And then class... |
2019/01/14 | 967 | 4,076 | <issue_start>username_0: How do you distinguish between a complex and a simple model in machine learning? Which parameters control the complexity or simplicity of a model? Is it the number of inputs, or maybe the number of layers?
Moreover, when should a simple model be used instead of a complex one, and vice-versa?<i... |
2019/01/14 | 780 | 3,272 | <issue_start>username_0: I was thinking of something of the sort:
1. Build a program (call this one fake user) that generates lots and lots and lots of data based on the usage of another program (call this one target) using stimuli and response. For example, if the target is a minesweeper, the fake user would play the... |
2019/01/14 | 3,191 | 10,187 | <issue_start>username_0: I was trying to understand the loss function of GANs, but I found a little mismatch between different papers.
This is taken from [the original GAN paper](https://arxiv.org/pdf/1406.2661.pdf):
>
> The adversarial modeling framework is most straightforward to apply when the models are both mul... |
2019/01/15 | 649 | 2,874 | <issue_start>username_0: Is there any way to control the extraction of features? How do I determine which features are been learned during training, i.e relevant information is been learned or not?<issue_comment>username_1: There are methods called "scoring systems" where you give a image scores such as "0.9 stripes, 0... |
2019/01/16 | 1,018 | 4,297 | <issue_start>username_0: What is "bad local minima"?
The following papers all mention this expression.
1. [Eliminating all bad Local Minima from Loss Landscapes without even adding an Extra Unit](https://arxiv.org/abs/1901.03909)
2. [limination of All Bad Local Minima in Deep Learning](https://arxiv.org/abs/1901.002... |
2019/01/16 | 386 | 1,520 | <issue_start>username_0: For example, I have the following csv: [training.csv](https://gist.github.com/JafferWilson/3ab8ee88f3fc32e78579a1054aac757d)
I want to know how I can determine which column will be the best feature for getting the output prediction before I go for machine training.
Please do share your re... |
2019/01/16 | 1,867 | 8,327 | <issue_start>username_0: There are many people trying to show how neural networks are still very different from humans, but I fail to see in what way human brains are different from neural models in anything but complexity.
The way we learn is similar, the way we process information is similar, the ways we predict out... |
2019/01/17 | 901 | 3,447 | <issue_start>username_0: I've seen the Monte Carlo return $G\_{t}$ being used in REINFORCE and the TD($0$) target $r\_t + \gamma Q(s', a')$ in vanilla actor-critic. However, I've never seen someone use the lambda return $G^{\lambda}\_{t}$ in these situations, nor in any other algorithms.
Is there a specific reason for... |
2019/01/18 | 1,203 | 4,747 | <issue_start>username_0: I am working to build a reinforcement agent with DQN. The agent would be able to place buy and sell orders for a day trading purpose. I am facing a little problem with that project. The question is "how to tell the agent to maximize the profit and avoid the transaction where the profit is less ... |
2019/01/19 | 3,717 | 15,067 | <issue_start>username_0: I have an electromagnetic sensor and electromagnetic field emitter.
The sensor will read power from the emitter. I want to predict the position of the sensor using the reading.
Let me simplify the problem, suppose the sensor and the emitter are in 1 dimension world where there are only positio... |
2019/01/22 | 361 | 1,326 | <issue_start>username_0: [](https://i.stack.imgur.com/thmqC.png)
They only reference in the paper that the position embeddings are learned, which is different from what was done in ELMo.
ELMo paper - <https://arxiv.org/pdf/1802.05365.pdf>
BERT paper... |
2019/01/24 | 2,043 | 9,158 | <issue_start>username_0: Imagine that I have an artificial neural network with a single hidden layer and that I am using ReLU as my activating function.
If by change I initialize my bias and my weights in such a form that:
$$
X \* W + B < 0
$$
for every input **x** in **X** then the partial derivate of the Loss functio... |
2019/01/24 | 817 | 3,075 | <issue_start>username_0: I am curious if it is possible to do so.
For example, if I supply
* $[0, 1, 2, 3, 4, 5]$, the model should return "natural number sequence",
* $[1,3,5,7,9,11]$, it should return "natural number with step of $2$",
* $[1,1,2,3,5]$, it should return "Fibonacci numbers",
* $[1,4,9,16,25]$, it sho... |
2019/01/25 | 1,737 | 7,466 | <issue_start>username_0: I'm quite new to the field of computer vision and was wondering what are the purposes of having the boundary boxes in object detection.
Obviously, it shows where the detected object is, and using a classifier can only classify one object per image, but my question is that
1. If I don't need t... |
2019/01/25 | 1,443 | 5,638 | <issue_start>username_0: I'm reading the book *Artificial Intelligence: A Modern Approach* (by <NAME> and <NAME>).
However, I don't understand the difference between search and planning. I was more confused when I saw that some search problems can be determined in a planning way. My professor explained to me in a conf... |
2019/01/24 | 1,086 | 4,418 | <issue_start>username_0: In [Open AI's actor-critic](https://github.com/pytorch/examples/blob/master/reinforcement_learning/actor_critic.py) and in [Open AI's REINFORCE](https://github.com/pytorch/examples/blob/master/reinforcement_learning/reinforce.py), the rewards are being normalized like so
```
rewards = (rewards... |
2019/01/26 | 1,024 | 4,466 | <issue_start>username_0: I am trying to make a personal ML project where my objective is using a photo from an invoice, for instance, a Walmart invoice, classify it as being a Walmart invoice and extract the total amount spent. I would then save this information in a relational database and infer some statistics about ... |
2019/01/26 | 449 | 2,043 | <issue_start>username_0: I am using [Open AI's](https://github.com/pytorch/examples/blob/master/reinforcement_learning/actor_critic.py) code to do a RL task on an environment that I built myself.
I tried some network architectures, and they all converge, faster or slower on CartPole.
On my environment, the reward see... |
2019/01/27 | 611 | 2,517 | <issue_start>username_0: In an MLP with ReLU activation functions after each hidden layer (except the final),
Let's say the final layer should output positive and negative values.
With ReLU intermediary activations, this is still possible because the final layer, despite taking in positive inputs only, can combine th... |
2019/01/28 | 883 | 4,095 | <issue_start>username_0: Can someone explain what is the process of learning? What does it mean to learn something?<issue_comment>username_1: That is a wonderfully fundamental question.
>
> Learning is the use of a system to change another system so that, instead of doing what it did before, which may have been nothi... |
2019/01/29 | 1,536 | 6,268 | <issue_start>username_0: By reading the abstract of [Neural Networks and Statistical Models](https://people.orie.cornell.edu/davidr/or474/nn_sas.pdf) paper it would seem that ANNs are statistical models.
In contrast [Machine Learning is not just glorified Statistics](https://towardsdatascience.com/no-machine-learning-... |
2019/01/29 | 939 | 3,701 | <issue_start>username_0: I'm trying to train a neural network on evaluating chess positions if rather white (0.0) or black would win (1.0)
Currently the input consists of 4 bits per chess field (piece id 0 - 12, equals 64\*4). Factors like castling are being ignored for now. Also, all training sets are random position... |
2019/01/30 | 1,654 | 7,040 | <issue_start>username_0: From what I understand, Monte Carlo Tree Search Algorithm is a solution algorithm for model free reinforcement learning (RL).
Model free RL means agent doesnt know the transition and reward model. Thus for it to know which next state it will observe and next reward it will get is for the agent... |
2019/01/30 | 1,860 | 7,382 | <issue_start>username_0: ***Status***:
For a few weeks now, I have been working on a Double DQN agent for the `PongDeterministic-v4` environment, which you can find [here](https://github.com/hridayns/Reinforcement-Learning/tree/master/Atari).
A single training run lasts for about 7-8 million timesteps (about 7000 epi... |
2019/01/31 | 637 | 2,645 | <issue_start>username_0: I'm doing bachaleor thesis on traffic sign detection using single shot detector called YOLO. These single shot detectors can perform detection of objects in image and so they have specific way of training, ie. training on full images. Thats quite problem for me, because the biggest real dataset... |
2019/01/31 | 684 | 2,722 | <issue_start>username_0: I recently came across a paper on [Deep Ranking](https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/42945.pdf). I was wondering whether this could be used to classify book covers as book titles. (For example, if I had a picture for the cover of the second HP book, t... |
2019/02/01 | 725 | 2,911 | <issue_start>username_0: I am working on a project where the Neural Network weights must be quantized on 8 or 16 bits for an embedded platform, thus I will lose some precision.
Since our platform does not have floating point arithmetic we need to quantize the weights. By quantizing i mean taking the max absolute value... |
2019/02/01 | 574 | 2,235 | <issue_start>username_0: Suppose we have a data set consists of columns
>
> TransactionId, CardNo, TransactionDate
>
>
>
then how can we calculate the customer purchase interval (means if customer A purchased on Jan 1st and after 10 days he again purchased, and then he again purchased after 15 days.) and how to p... |
2019/02/01 | 1,950 | 6,685 | <issue_start>username_0: In [experience replay](https://datascience.stackexchange.com/q/20535/10640), the update rule follows the loss:
>
> $$
> L\_i(\theta\_i) = \mathbb{E}\_{(s\_t, a\_t, r\_t, s\_{t+1}) \sim U(D)} \left[ \left(r\_t + \gamma \max\_{a\_{t+1}} Q(s\_{t+1}, a\_{t+1}; \theta\_i^-) - Q(s\_t, a\_t; \theta\... |
2019/02/02 | 946 | 3,484 | <issue_start>username_0: I am reading [BayesChess: A computer chess program based on Bayesian networks](https://core.ac.uk/download/pdf/143454908.pdf) (<NAME>; 2008)
It is a chess-playing engine using Bayesian networks. The following is mentioned about the heuristic function in section 3.
>
> Here the heuristic is d... |
2019/02/02 | 1,281 | 4,609 | <issue_start>username_0: In my country, the Expert System class is mandatory, if you want to take the AI specialization in most universities. In class, I learned how to make a rule-based system, [forward chaining](https://en.wikipedia.org/wiki/Forward_chaining), [backward chaining](https://en.wikipedia.org/wiki/Backwar... |
2019/02/05 | 1,239 | 4,532 | <issue_start>username_0: In a neural network, each neuron represents some part of the input. For example, in the case of a MNIST digit, consider the stem of the number 9. Each neuron in the NN represents some part of this digit.
1. What determines which neuron will represent which part of the digit?
2. Is it possible ... |
2019/02/07 | 715 | 3,136 | <issue_start>username_0: I have two Machine Learning models (I use LSTM) that have a different result on the validation set (~100 samples data):
* Model A: Accuracy: ~91%, Loss: ~0.01
* Model B: Accuracy: ~83%, Loss: ~0.003
The size and the speed of both models are almost the same. So, which model should I choose?<is... |
2019/02/07 | 2,315 | 8,509 | <issue_start>username_0: In this [video](https://www.youtube.com/watch?v=Jw3ZnWFjDfM), the lecturer states that $R(s)$, $R(s, a)$ and $R(s, a, s')$ are equivalent representations of the reward function. Intuitively, this is the case, according to the same lecturer, because $s$ can be made to represent the state and the... |
2019/02/07 | 813 | 3,291 | <issue_start>username_0: I have a dataset which I have loaded as a data frame in Python. It consists of 21392 rows (the data instances, each row is one sample) and 1972 columns (the features). The last column i.e. column 1972 has string type labels (14 different categories of target labels). I would like to use a CNN t... |
2019/02/09 | 1,825 | 6,210 | <issue_start>username_0: I am studying a knowledge base (KB) from the book "Artificial Intelligence: A Modern Approach" (by <NAME> and <NAME>) and from [this series of slides](https://www.cs.princeton.edu/courses/archive/fall16/cos402/lectures/402-lec12.pdf).
A formula is satisfiable if there is some assignment to the... |
2019/02/11 | 547 | 1,469 | <issue_start>username_0: In the diagram below, there are three variables: `X3` is a function of (depends on) `X1` and `X2`, `X2` also depends on `X1`. More specifically, `X3 = f(X1, x2)` and `X2 = g(X1)`. Therefore, `X3 = f(X1, g(X1))`.
[](https://i.s... |
2019/02/11 | 459 | 1,377 | <issue_start>username_0: What is the difference between automatic transcription and automatic speech recognition? Are they the same?
Is my following interpretation correct?
*Automatic transcription*: it converts the speech to text by looking at the whole spoken input
*Automatic speech recognition*: it converts the ... |
2019/02/12 | 1,848 | 7,205 | <issue_start>username_0: From [the reinforcement learning book section 13.3](http://incompleteideas.net/book/bookdraft2017nov5.pdf):
[](https://i.stack.imgur.com/Ex4a5.png)
Using pytorch, I need to calculate a loss, and then the gradient is calculate... |
2019/02/12 | 639 | 2,621 | <issue_start>username_0: I am a beginner, just started studying around NLP, specifically various language models. So far, my understanding is that - the goal is to understand/produce natural language.
So far the methods I have studied speak about correlation of words, using correct combination to make a meaningful sen... |
2019/02/13 | 258 | 1,214 | <issue_start>username_0: For example, you train on dataset 1 with an adaptive optimizer like Adam. Should you reload the learning schedule, etc., from the end of training on dataset 1 when attempting transfer to dataset 2? Why or why not?<issue_comment>username_1: When doing transfer learning it makes sense to have dif... |
2019/02/13 | 681 | 2,740 | <issue_start>username_0: I am purchasing Titan RTX GPU. Everything seems fine with that except float32 & float64 performance which seems lower vis-a-vis some of its counter parts. I wanted to understand if single precision and double precision performance of GPU affect deep learning training or efficiency ? We work mos... |
2019/02/13 | 2,284 | 6,254 | <issue_start>username_0: In the paper [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://arxiv.org/pdf/1801.01290.pdf), they define the loss function for the policy network as
$$
J\_\pi(\phi)=\mathbb E\_{s\_t\sim \mathcal D}\left[D\_{KL}\left(\pi\_\phi(\cdot|s\_... |
2019/02/14 | 1,664 | 5,205 | <issue_start>username_0: In reinforcement learning, we often define two functions, the *state-value function*
$$V^\pi(s) = \mathbb{E}\_{\pi} \left[\sum\_{k=0}^{\infty}
\gamma^{k}R\_{t+k+1} \Bigg| S\_t=s \right]$$
and the *state-action-value function*
$$Q^\pi(s,a) = \mathbb{E}\_{\pi}\left[\sum\_{k=0}^{\infty} \gamma... |
2019/02/15 | 1,077 | 4,462 | <issue_start>username_0: Is the optimal policy always stochastic (that is, a map from states to a probability distribution over actions) if the environment is also stochastic?
Intuitively, if the environment is *deterministic* (that is, if the agent is in a state $s$ and takes action $a$, then the next state $s'$ is ... |
2019/02/15 | 2,189 | 9,048 | <issue_start>username_0: This is not meant to be negative or a joke but rather looking for a productive solution on AI development, engineering and its impact on human life:
Lately with my Google searches, the AI model keeps auto filling the ending of my searches with:
“...in Vietnamese”
And
“...in a Vietnamese ho... |
2019/02/16 | 1,898 | 8,385 | <issue_start>username_0: This is an excerpt taken from [Sutton and Barto](https://web.stanford.edu/class/psych209/Readings/SuttonBartoIPRLBook2ndEd.pdf) (pg. 3):
>
> Another key feature of reinforcement learning is that it explicitly considers the whole problem of a goal-directed agent interacting with an uncertain e... |
2019/02/16 | 417 | 1,759 | <issue_start>username_0: I am trying to understand how RNNs are used for sequence modelling.
On a [tutorial here](https://blog.keras.io/a-ten-minute-introduction-to-sequence-to-sequence-learning-in-keras.html), it mentions that if you want to translate say a sentence from English to French you can use an encoder-decod... |
2019/02/16 | 420 | 1,922 | <issue_start>username_0: In reinforcement learning (RL), there are model-based and model-free algorithms. In short, model-based algorithms use a transition model $p(s' \mid s, a)$ and the reward function $r(s, a)$, even though they do not necessarily compute (or estimate) them. On the other hand, model-free algorithms ... |
2019/02/16 | 1,974 | 8,094 | <issue_start>username_0: What is self-supervised learning in machine learning? How is it different from supervised learning?<issue_comment>username_1: Introduction
------------
The term *self-supervised learning* (SSL) has been used (sometimes differently) in different contexts and fields, such as representation learn... |
2019/02/17 | 1,021 | 3,948 | <issue_start>username_0: I'm working on my own implementation of NEAT algorithm based on the original 2002 paper called "[Efficient Reinforcement Learning through Evolving Neural Network Topologies](http://nn.cs.utexas.edu/downloads/papers/stanley.gecco02_1.pdf)" (by <NAME> and <NAME>). The way the algorithm is designe... |
2019/02/17 | 3,079 | 13,074 | <issue_start>username_0: The "AI Singularity" or "[Technological Singularity](https://en.wikipedia.org/wiki/Technological_singularity)" is a vague term that roughly seems to refer to the idea of:
1. Humans can design algorithms
2. Humans can improve algorithms
3. Eventually algorithms we design might end up being as g... |
2019/02/18 | 695 | 2,873 | <issue_start>username_0: I am generating images that consist of points, where the object's location is where the most overlap of points occurs.
[](https://i.stack.imgur.com/7FyHl.png)
In this example, the object location is $(25, 51)$.
I am trying to tra... |
2019/02/18 | 505 | 1,710 | <issue_start>username_0: I asked a question a while ago here and since then I've been solving the issues within my code but I have just one question... This is the formula for updating the Q-Matrix in Q-Learning:
$$Q(s\_t, a\_t) = Q(s\_t, a\_t) + \alpha \times (R+Q(s\_{t+1}, max\_a)-Q(s\_t, a\_t))$$
However, I saw a ... |
2019/02/18 | 976 | 4,026 | <issue_start>username_0: In a neural network, the number of neurons in the hidden layer corresponds to the complexity of the model generated to map the inputs to output(s). More neurons creates a more complex function (and thus the ability to model more nuanced decision barriers) than a hidden layer with less nodes.
B... |
2019/02/18 | 1,259 | 5,564 | <issue_start>username_0: In some newer robotics literature, the term system identification is used in a certain meaning. The idea is not to use a fixed model, but to create the model on the fly. So it is equal to a model-free system identification. Perhaps a short remark for all, who doesn't know what the idea is. Syst... |
2019/02/18 | 766 | 3,377 | <issue_start>username_0: Fuzzy logic is typically used in control theory and engineering applications, but is it connected fundamentally to classification systems?
Once I have a trained neural network (multiple inputs, one output), I have a nonlinear function that will turn a set of inputs into a number that will esti... |
2019/02/19 | 517 | 2,106 | <issue_start>username_0: Some (stock market) traders have the ability to produce a high percentage of winning trades (80%+, positive return) over years. I had the chance to look into real money trades of two such traders and I also got trading instructions from them for research.
Now the interesting part is that if y... |
2019/02/20 | 430 | 1,667 | <issue_start>username_0: I trained a neural network on the [UNSW-NB15 dataset](https://www.unsw.adfa.edu.au/unsw-canberra-cyber/cybersecurity/ADFA-NB15-Datasets/), but, during training, I am getting spikes in the loss function. The algorithms see part of this UNSW dataset a single time. The loss function is plotted aft... |
2019/02/20 | 1,716 | 6,880 | <issue_start>username_0: This question is related to [What does "stationary" mean in the context of reinforcement learning?](https://ai.stackexchange.com/q/7640/2444), but I have a more specific question to clarify the difference between a non-stationary policy and a state that includes time.
My understanding is that,... |
2019/02/20 | 769 | 3,167 | <issue_start>username_0: I am confused about how neural networks weigh different features or inputs.
Consider this example. I have 3 features/inputs: an image, a dollar amount, and a rating. However, since one feature is an image, I need to represent it with very high dimensionality, for example with $128 \times 128 =... |
2019/02/21 | 638 | 2,611 | <issue_start>username_0: How to detect liveness of face using face landmark points?
I am getting face landmarks from android camera frames. And I want to detect liveness using these landmark points.
How to tell if a human is making a specific movement that can be useful for liveness detection?<issue_comment>username_... |
2019/02/22 | 1,581 | 7,152 | <issue_start>username_0: I've been looking at various bounding box algorithms, like the three versions of RCNN, SSD and YOLO, and I have noticed that not even the original papers include pseudocode for their algorithms. I have built a CNN classifier and I am attempting to incorporate bounding box regression, though I a... |
2019/02/22 | 1,787 | 6,464 | <issue_start>username_0: I came across these 2 algorithms, but I cannot understand the difference between these 2, both in terms of implementation as well as intuitionally.
So, what difference does the second point in both the slides refer to?
[](htt... |
2019/02/23 | 3,688 | 14,148 | <issue_start>username_0: In reinforcement learning, successive states (actions and rewards) can be correlated. An [experience replay](http://www.incompleteideas.net/lin-92.pdf) buffer was used, in the [DQN architecture](https://web.stanford.edu/class/psych209/Readings/MnihEtAlHassibis15NatureControlDeepRL.pdf), to avoi... |
2019/02/26 | 280 | 1,006 | <issue_start>username_0: I have just found the paper and documentation about GAN 2.0, the new face creator from Nvidia.
On the website <https://thispersondoesnotexist.com/> they have used this approach to create realistic faces. Unfortunately, the website does not exist anymore.
Is there another webpage demonstrating... |
2019/02/27 | 711 | 2,980 | <issue_start>username_0: I have already implemented a relatively simple DQN on Pacman.
Now I would like to clearly understand the difference between a DQN and the techniques used by AlphaGo zero/AlphaZero and I couldn't find a place where the features of both approaches are compared.
Also sometimes, when reading thr... |
2019/02/27 | 839 | 3,179 | <issue_start>username_0: In the context of Deep Q Network, a target network is usually utilized. The target network is a slow changing network with a changing rate as its hyperparameter. This includes both replacement update every $N$ iterations and slowly update every iteration.
Since the rate is hard to fine tune ma... |
2019/02/27 | 446 | 1,507 | <issue_start>username_0: I am trying to solve part b of the exercise 3.6 (page 113) from the book [Artificial Intelligence: A Modern Approach](https://faculty.psau.edu.sa/filedownload/doc-7-pdf-a154ffbcec538a4161a406abf62f5b76-original.pdf).
More specifically, I need to give a complete problem formulation (that is pre... |
2019/02/27 | 599 | 2,234 | <issue_start>username_0: Suppose I want to build a neural network regression model that takes one input and return one output.
Here's the training data:
```
0.1 => 0.1
0.2 => 0.2
0.1 => -0.1
```
You will see that there are 2 inputs `0.1` that matches to different output values `0.1` and `-0.1`. So what will happen ... |
2019/02/27 | 409 | 1,590 | <issue_start>username_0: I hope this question is not too broad or general. I have a very large set of images all of which contain text (some have more, some less). All of them have been tagged as containing, say, English text or Korean. I wonder if convolutional neural networks would be a good approach to classify thes... |
2019/02/28 | 1,903 | 3,939 | <issue_start>username_0: I have a bayesian network, which has the following data:
$P(S) = 0.07$
$P(A) = 0.01$
$P(F \mid S,A) = 1.0$
$P(F \mid S, \lnot A) = 0.7$
$P(F \mid \lnot S, A) = 0.9$
$P(F \mid \lnot S, \lnot A) = 0.1$
And I'm asked to get $P(F \mid S)$. Is it possible? How can I deduce it?<issue_comment>... |
2019/03/01 | 369 | 1,414 | <issue_start>username_0: I was reading an article on [Medium](https://chatbotsjournal.com/why-i-should-should-not-invest-in-chatbot-df952cbabf10) and wanted to make it clear whether a bot created on IBM Watson is an intelligent one or unintelligent.
>
> Simply put, there are 2 types of chatbots — unintelligent ones t... |
2019/03/01 | 1,726 | 7,048 | <issue_start>username_0: The following plot shows error function output based on system weights.
Two equal local minima are shown in green pointers. Note that the red dots are not related to the question.
Does the right one generalize better compared to the left one?
My assumption is that for the right minimum if th... |
2019/03/02 | 792 | 3,059 | <issue_start>username_0: I just read about the concept of a [parse tree](https://en.wikipedia.org/wiki/Parse_tree).
In my understanding, a valid parse tree of a sentence needs to be validated by a linguistic expert. So, I concluded, a sentence only has one parse tree.
But, is that correct? Is it possible a sentence h... |
2019/03/03 | 702 | 2,870 | <issue_start>username_0: So I was wondering, why I have only encountered square loss function also known as MSE. The only nice property of MSE I am so far aware of is its convex nature. But then all equations of the form $x^{2n}$ where $n$ is an integer belongs to the same family.
My question is what makes MSE the mos... |
2019/03/04 | 744 | 3,029 | <issue_start>username_0: The (discrete and continuous) Fourier transform (FT) is used in signal processing in order to convert a signal (or function) in a certain domain (e.g. the time domain) to another domain (e.g., frequency domain). There are several resources on the web that attempt to explain the FT at different ... |
2019/03/05 | 491 | 1,920 | <issue_start>username_0: It is possible that the view of what is impressive enough in computer behavior to be called intelligence changes with each decade as we adjust to what capabilities are made available in products and services.<issue_comment>username_1: My sense is that they would, based on a high-level take of [... |
2019/03/05 | 1,769 | 6,223 | <issue_start>username_0: I wanted to clarify the term 'acting greedily'. What does it mean? Does it correspond to the immediate reward, future reward or both combined? I want to know the actions that will be taken in 2 cases:
* $v\_\pi(s)$ is known and $R\_s$ is also known (only).
* $q\_{\pi}(s, a)$ is known and $R\_s... |
2019/03/05 | 267 | 1,025 | <issue_start>username_0: I'm looking to build from scratch an implementation of the [wake-sleep algorithm](https://www.cs.toronto.edu/~hinton/absps/ws.pdf), for unsupervised learning with neural networks. I plan on doing this in Python in order to better understand how it works. In order to facilitate my task, I was wo... |
2019/03/07 | 652 | 2,794 | <issue_start>username_0: I'm studying how SPP (Spatial, Pyramid, Pooling) works. SPP was invented to tackle the fix input image size in CNN. According to the original paper <https://arxiv.org/pdf/1406.4729.pdf>, the authors say:
>
> convolutional layers do not require a fixed image size and can
> generate feature m... |
2019/03/07 | 577 | 2,533 | <issue_start>username_0: I have series of sensors (around 4k) and each sensor will measure the amplitudes at each point.Suppose I train the neural network with sufficent set of 4k values (N \* 4k shape). The machine will find a pattern in the series of values.If the values stray away from the pattern (that is anomaly) ... |
2019/03/07 | 595 | 1,998 | <issue_start>username_0: >
> Excercise 3.5 The equastions in Section 3.1 are for the continuing
> case and need to be modified (very slightly) to apply to episodic
> tasks. Show that you know the modifications needed by giving the
> modified version of (3.3).
>
>
>
$\displaystyle\sum\_{s^{\prime} \in S} \displa... |
2019/03/08 | 329 | 1,408 | <issue_start>username_0: I am working on a MLP neural networks, using supervised learning (2 classes and multi-class classification problems). For the hidden layers, I am using $\tanh$ (which produces an output in the range $[-1, 1]$) and for the output layer a softmax (which gives the probability distribution between ... |
2019/03/08 | 239 | 1,043 | <issue_start>username_0: Is there a way to run C64, Nintendo Gameboy, or other platforms in an OpenAI's Gym-like environment? It seems that we can run only Atari games.<issue_comment>username_1: For this particular classification problem, I would recommend you using a softmax function whose output range is [0,1].
Th... |
2019/03/12 | 3,005 | 11,906 | <issue_start>username_0: What is geometric deep learning (GDL)?
Here are a few sub-questions
* How is it different from deep learning?
* Why do we need GDL?
* What are some applications of GDL?<issue_comment>username_1: The article [Geometric deep learning: going beyond Euclidean data](https://arxiv.org/abs/1611.0809... |
2019/03/13 | 3,043 | 12,138 | <issue_start>username_0: I am making an AI model to predict monthly retail sales of a motor cycle spare parts shop, for that to be possible I have to first create a dataset. The problem I am facing is what features should the dataset have?
I already did some research on some other datasets but still I want to know spe... |
2019/03/14 | 3,019 | 12,012 | <issue_start>username_0: I need to create model which will find suspicious entries or anomalies in a network, whose characteristics or features are the `asset_id`, `user_id`, IP accessed from and `time_stamp`.
Which unsupervised anomaly detection algorithms or models should I use to solve this task?<issue_comment>user... |
2019/03/14 | 441 | 1,597 | <issue_start>username_0: I read top articles on Google Search about Deep Q-Learning:
* <https://medium.freecodecamp.org/an-introduction-to-deep-q-learning-lets-play-doom-54d02d8017d8>
* <https://skymind.ai/wiki/deep-reinforcement-learning>
* <https://neuro.cs.ut.ee/demystifying-deep-reinforcement-learning/>
* [Q-Learn... |
2019/03/14 | 191 | 785 | <issue_start>username_0: I am not new to AI and did some work for few months but completely new to text to audio. Yes I used text to audio tools a decade back... but I would like to know where exactly we stand in terms of text to audio today.
I already did some research, it seems like traditional way of text to audio ... |
2019/03/14 | 2,960 | 11,530 | <issue_start>username_0: What is non-Euclidean data?
Here are some sub-questions
* Where does this type of data arise? I have come across this term in the context of geometric deep learning and graph neural networks.
* Apparently, graphs and manifolds are non-Euclidean data. Why exactly is that the case?
* What is th... |
2019/03/14 | 1,786 | 5,315 | <issue_start>username_0: In certain reinforcement learning (RL) proofs, the operators involved are assumed to be *non-expansive*. For example, on page 6 of the paper [Generalized Markov Decision Processes: Dynamic-programming and Reinforcement-learning Algorithms](http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.... |
2019/03/17 | 1,823 | 7,843 | <issue_start>username_0: In general, the word "latent" means "hidden" and "to embed" means "to incorporate". In machine learning, the expressions "hidden (or latent) space" and "embedding space" occur in several contexts. More specifically, an [embedding](https://en.wikipedia.org/wiki/Word_embedding) can refer to a vec... |
2019/03/17 | 333 | 1,474 | <issue_start>username_0: I am trying to understand how weights are actually gotten. What is generating them in a neural network? What is the algorithm that gives them certain values?<issue_comment>username_1: Typically, weights are randomly initialized. Then, as the model is [optimized](https://en.wikipedia.org/wiki/Ar... |
2019/03/20 | 828 | 3,824 | <issue_start>username_0: If we have classified 1000 people's faces; how do we ensure the network tells us when it encounters a new person?<issue_comment>username_1: You can't really make the network tell you if a face is new (unless you actually train a network with that particular purpose, maybe). Ideally, if you feed... |
2019/03/20 | 1,012 | 4,458 | <issue_start>username_0: I am fine-tuning a VGG16 model on 20 classes with 500k images I was wondering how do you chose the size of the dense layer (the one before the prediction layer which has a size 20). I would prefer not to do a grid search seeing how long it take to train my model.
Also how many Dense layer shou... |
2019/03/20 | 1,022 | 4,519 | <issue_start>username_0: So i have been playing around with `neat-python`. I made a program, applying neat, to play pinball on the Atari 2600. The code for that can be found in the file `test2.py` [here](https://github.com/niallmandal/atari_pinball_tutorial)
Now based on that, I would like to do the same, but on a 2 p... |
2019/03/20 | 1,187 | 4,511 | <issue_start>username_0: I have read a lot about RL algorithms, that update the action-value function at each step with the currently gained reward. The requirement here is, that the reward is obtained after each step.
I have a case, where I have three steps, that have to be passed in a specific order. At each step th... |
2019/03/21 | 1,311 | 5,440 | <issue_start>username_0: I want to solve a problem using Reinforcement Learning on a 20x20 board. An agent (a mouse) has to get the highest possible rewards as fast as possible by collecting cheese, which there are 10 in total. The agent has a fixed amount of time for solving this problem, namely 500 steps per game. Th... |