{"id": "2604.06464", "title": "Weighted Bayesian Conformal Prediction", "abstract": "Conformal prediction provides distribution-free prediction intervals with finite-sample coverage guarantees, and recent work by Snell \\& Griffiths reframes it as Bayesian Quadrature (BQ-CP), yielding powerful data-conditional guarantees via Dirichlet posteriors over thresholds. However, BQ-CP fundamentally requires the i.i.d. assumption. Meanwhile, weighted conformal prediction handles distribution shift via importance weights but remains frequentist, producing only point-estimate thresholds. We propose \\textbf{Weighted Bayesian Conformal Prediction (WBCP)}, which generalizes BQ-CP to arbitrary importance-weighted settings by replacing the uniform Dirichlet $\\Dir(1,\\ldots,1)$ with a weighted Dirichlet $\\Dir(\\neff \\cdot \\tilde{w}_1, \\ldots, \\neff \\cdot \\tilde{w}_n)$, where $\\neff$ is Kish's effective sample size. We prove four theoretical results: (1)~$\\neff$ is the unique concentration parameter matching frequentist and Bayesian variances; (2)~posterior standard deviation decays as $O(1/\\sqrt{\\neff})$; (3)~BQ-CP's stochastic dominance guarantee extends to per-weight-profile data-conditional guarantees; (4)~the HPD threshold provides $O(1/\\sqrt{\\neff})$ improvement in conditional coverage. We instantiate WBCP for spatial prediction as \\emph{Geographical BQ-CP}, where kernel-based spatial weights yield per-location posteriors with interpretable diagnostics. Experiments on synthetic and real-world spatial datasets demonstrate that WBCP maintains coverage guarantees while providing substantially richer uncertainty information.", "date": "2026-04-07", "categories": []} {"id": "2606.20115", "title": "When Calibration Fails the Vulnerable Hospital: Federated Conformal Risk Control via Risk-Curve Shrinkage", "abstract": "Conformal risk control (CRC) provides distribution-free guarantees on segmentation quality by calibrating a prediction-set threshold on held-out data. In federated deployments, the standard approach pools calibration scores across sites into a single threshold. We provide the first quantification, on real multi-institutional brain tumor data (FeTS-2022, 1,251 subjects, 20 institutions), showing that this naive pooled CRC protects the average hospital but violates coverage at 40% of individual institutions, with the worst site exceeding the target false-negative rate by 7.8 percentage points. The naive alternative, per-site local CRC, largely restores coverage but inflates prediction sets by 83x, rendering them clinically useless. We propose a shrinkage-based federated CRC protocol: each site transmits only its empirical risk curve (G scalars) to a server, which computes a shrinkage-regularized threshold per site. A single hyperparameter n0 smoothly trades worst-case coverage for prediction-set efficiency; leave-one-site-out sensitivity analysis identifies n0=19, achieving 2.7/20 violations at 2.0x stretch. We further show that direct Lagrangian optimization of coverage budgets fails, concentrating risk on vulnerable hospitals, and that the finite-sample correction term is essential: removing it triples violations. The marginal CRC guarantee is preserved by construction under the stated site-mixture assumption; per-site coverage is validated across four targets with three seeds. No patient-level images, masks, or per-volume scores leave any site.", "date": "2026-06-18", "categories": []} {"id": "2605.20726", "title": "Everywhere Valid Bounds on False Discovery Proportions in Conformal Inference", "abstract": "Modern applications of conformal inference to multiple testing problems, such as outlier detection and candidate selection, often involve selecting test samples whose conformal p-values fall below a threshold. The quality of such methods is often measured by the false discovery proportion (FDP), defined as the fraction of incorrect selections. Existing approaches typically control the expected value of the FDP, using methods such as the Benjamini-Hochberg procedure. This approach fails to provide high-probability bounds on the realized false discovery proportion and invalidates statistical guarantees if the rejection threshold is selected after inspecting the data. This paper establishes finite-sample, distribution-free upper bounds on the FDP that hold simultaneously over all possible rejection thresholds, enabling arbitrary post hoc selection of the threshold. Simultaneous validity is achieved by constructing a high-probability envelope for the empirical distribution function of null conformal p-values by sampling from their joint distribution. Furthermore, our framework allows practitioners to modulate the envelope's shape, thereby producing tight bounds in rejection regions of primary interest. We use this flexible approach to derive simultaneous FDP upper bounds for both outlier detection and conformal selection. We demonstrate through synthetic and real-data experiments that the resulting bounds are both valid and substantially less conservative than those derived from existing approaches.", "date": "2026-05-20", "categories": []} {"id": "2601.21455", "title": "Questioning the Coverage-Length Metric in Conformal Prediction: When Shorter Intervals Are Not Better", "abstract": "Conformal prediction(CP) has become a cornerstone of distribution-free uncertainty quantification, conventionally evaluated by its coverage and interval length. This work critically examines the sufficiency of these standard metrics. We demonstrate that the interval length might be deceptively improved through a counter-intuitive approach termed Prejudicial Trick(PT), while the coverage remains valid. Specifically, for any given test sample, PT probabilistically returns an interval, which is either null or constructed using an adjusted confidence level, thereby preserving marginal coverage. While PT potentially yields a deceptively lower interval length, it introduces practical vulnerabilities: the same input can yield completely different prediction intervals across repeated runs of the algorithm. We formally derive the conditions under which PT achieves these misleading improvements and provide extensive empirical evidence across various regression and classification tasks. Furthermore, we introduce a new metric interval stability which helps detect whether a new CP method implicitly improves the length based on such PT-like techniques. Code is available at https://github.com/benben-cd/PT-Conformal-Prediction.", "date": "2026-01-29", "categories": []} {"id": "2606.14909", "title": "Audited Conformal Prediction for Classification under Unknown Distribution Shift", "abstract": "We consider the problem of uncertainty quantification for a pretrained classification model deployed under unknown distribution shift. We propose Audited Conformal Prediction (ACP), a method that leverages a small labeled dataset from the target population to train an auxiliary audit model identifying inputs where the legacy model is likely to fail. By integrating the audit model's outputs into the conformal prediction framework, ACP produces prediction sets that guarantee marginal coverage while achieving substantially higher conditional coverage in practice than existing approaches. We develop and analyze two complementary integration strategies -- one targeting marginal coverage with improved conditional performance, the other providing explicit group-conditional coverage guarantees -- and establish theoretical guarantees for both. Experiments on synthetic and real-world datasets validate the method and illustrate trade-offs between prediction set size and conditional coverage.", "date": "2026-06-12", "categories": []} {"id": "2606.15217", "title": "Conformal Candidate Certification for Offline Model-Based Optimization", "abstract": "Offline model-based optimization (MBO) proposes candidates by optimizing a surrogate trained on a fixed historical dataset. Because candidates are deliberately out-of-distribution, surrogate rankings are least reliable exactly where the optimizer is most aggressive, yet existing methods provide no per-candidate statistical certificate that a design meets a target threshold. We propose \\emph{Conformal Candidate Certification} (CCC), a post-hoc wrapper that attaches a calibrated one-sided lower bound to each candidate and advances only those whose bound exceeds the target. We show that entropy-regularized surrogate maximization induces a Gibbs-tilted proposal, so the same surrogate supplies importance weights for weighted conformal prediction without a separate density-ratio estimation step. In a controlled synthetic study, CCC certifies $16.7\\%$ of an aggressive proposal pool with empirical coverage 0.990 at nominal 0.90, while standard conformal prediction ignoring the covariate shift collapses to 0.416 coverage.", "date": "2026-06-13", "categories": []} {"id": "2606.15950", "title": "Spectral Adaptive Conformal Prediction for Structured Non-Exchangeable Data", "abstract": "Conformal prediction gives prediction intervals with finite-sample coverage when the data are exchangeable. Many time-indexed datasets are not exchangeable. They have seasons, recurring regimes, changing frequencies, or other forms of structured dependence. This paper studies a simple way to use that structure. We propose spectral adaptive conformal prediction, a method that forms weighted conformal quantiles using local spectral similarity and then updates the target miscoverage level online. The spectral weights choose calibration residuals that look relevant to the current test point. The adaptive update corrects the long-run miss rate when uncertainty changes over time. We give an approximate coverage result for the fixed spectral weighted quantile and a deterministic long-run calibration result for the adaptive update. Simulations with recurring regimes and slowly changing frequencies, together with three U.S. real-data examples, show that the hybrid method can improve on fixed spectral weighting, while also showing that spectral weighting must be monitored through effective sample size diagnostics.", "date": "2026-06-14", "categories": []} {"id": "2606.15964", "title": "PromptShift-CRC: Drift-Aware Conformal Risk Control for Foundation Models Under Prompt and Domain Shift", "abstract": "Foundation models are now used in settings where the prompts they receive can change quickly. Users change, topics change, policies change, and the model may suddenly face a kind of request that was rare in the calibration data. This makes fixed calibration risky. Conformal prediction and conformal risk control give model-agnostic ways to control error, but they work best when the calibration data still look like the future data. This paper develops PromptShift CRC, a drift-aware conformal risk control method for foundation-model outputs under prompt and domain shift. The method embeds prompts and responses, measures how far the current prompt stream has moved from the calibration pool, gives more weight to relevant or recent calibration examples, and updates the risk level online after observed violations. It reports three practical diagnostics: realized risk error, prompt drift, and effective calibration size. We give conditions under which the method controls risk up to terms for distribution mismatch and weighted quantile uncertainty. In a synthetic prompt-shift benchmark, static conformal risk control fails sharply after drift, while PromptShift-CRC gives the best coverage among the adaptive baselines considered. We then evaluate the same calibration layer on public benchmark derived streams for question answering, toxicity, summarization factuality, and long-context hallucination risk", "date": "2026-06-14", "categories": []} {"id": "2606.16356", "title": "Simulation-Augmented Multi-Step Split Conformal Prediction for Aggregated Forecasts", "abstract": "We study uncertainty quantification for aggregated forecasting tasks such as annual totals and year-over-year growth rates. We propose SA-MSCP, a simulation-augmented multi-step split conformal method that generates future paths from cross-validated residuals using a block bootstrap and constructs prediction intervals from empirical quantiles. Experiments show that SA-MSCP improves empirical coverage over a simulated-path baseline for aggregated and growth-rate targets. Our results demonstrate that simulation-enhanced conformal calibration is an effective and general framework for uncertainty quantification in aggregated time-series forecasting.", "date": "2026-06-15", "categories": []} {"id": "2606.17014", "title": "Filtered Conformal Ellipsoids for Graph-Native Time Series", "abstract": "Joint prediction sets for multivariate time series should control a single event while adapting to cross-coordinate dependence. We study filtered conformal ellipsoids: a frozen state-space filter emits a one-step predictive mean and covariance, and split-conformal calibration is applied to the resulting Mahalanobis scores. The filter is used to choose the ellipsoid shape; conformal calibration chooses the scalar radius, so the construction benefits from a learned predictive covariance without relying on Gaussian tail probabilities for coverage. The main difficulty is that filtered scores are dependent and learned recurrent filters need not contract in their raw hidden state; we therefore analyse contraction in an observable predictive-law quotient that identifies hidden states producing the same future sequence of emitted Gaussian laws. Under a stable Bayes Gaussian-projection filter, covariance bounds, and a finite-horizon observability Fisher condition, small excess Gaussian negative log-likelihood implies contraction of the learned emitted laws. Combined with a threshold-autocovariance envelope this yields a Chebyshev-type approximate coverage bound for filtered split-conformal prediction under dependence; a sharper Bernstein-type bound requires an additional geometric-mixing concentration assumption. Under Gaussian oracle realisability we also obtain a near-oracle log-volume comparison within the class of conditionally valid Gaussian ellipsoid rules. We instantiate the framework with a GCN-GRU filter with diagonal-plus-low-rank covariance. On moderate-size graph-native traffic benchmarks (METRLA-$20$ and PEMSBAY-$50$), the learned filter gives sharper at-target ellipsoids than static-covariance and non-filter baselines; at full-graph scale and on non-graph-native datasets, factor and copula baselines can be stronger.", "date": "2026-06-15", "categories": []} {"id": "2402.16388", "title": "Leave-One-Out-, Bootstrap- and Cross-Conformal Anomaly Detectors", "abstract": "The need for uncertainty quantification in anomaly detection systems has become increasingly important. In this context, effectively controlling Type I error rates without inflating Type II error rates in these systems can build trust and reduce costs associated with false discoveries. The field of conformal anomaly detection emerges as a promising approach for providing respective statistical and finite-sample validity guarantees through model calibration. However, reliance on calibration data imposes practical limitations, especially in low-data regimes. In this work, we formally define and evaluate leave-one-out-, bootstrap-, and cross-conformal methods for conformal anomaly detection, building on methods from the field of conformal prediction. Looking beyond the classical split-conformal approach, we show that derived methods for calculating resampling-conformal $p$-values offer a practical compromise between the data efficiency of full-conformal (transductive) approaches and the computational efficiency of split-conformal (inductive) methods. We validate derived methods and quantify their improvements for a range of one-class classifiers and datasets.", "date": "2024-02-26", "categories": []} {"id": "2507.20068", "title": "PERRY: Policy Evaluation with Confidence Intervals using Auxiliary Data", "abstract": "Off-policy evaluation (OPE) methods estimate the value of a new reinforcement learning (RL) policy prior to deployment. Recent advances have shown that leveraging auxiliary datasets, such as those synthesized by generative models, can improve the accuracy of OPE methods. Unfortunately, such auxiliary datasets may also be biased, and existing methods for using data augmentation within OPE lack principled uncertainty quantification. In high stakes domains like healthcare, reliable uncertainty estimates are important for ensuring safe and informed deployment of RL policies. In this work, we propose two methods to construct valid confidence intervals for OPE with data augmentation. The first provides a confidence interval over $V^{\\pi}(s)$, the policy value conditioned on an initial state $s$. To do so we introduce a new conformal prediction method suitable for Markov Decision Processes (MDPs) with continuous state spaces, extending prior work to higher-dimensional settings. Second, we consider the more common task of estimating the average policy performance over all initial states, $V^{\\pi}$; we introduce a method that draws on ideas from doubly robust estimation and prediction powered inference. Across simulators spanning inventory management, robotics, healthcare, and a real healthcare dataset from MIMIC-IV, we find that our methods can effectively leverage auxiliary data and consistently produce confidence intervals that cover the ground truth policy values, unlike previously proposed methods. Our work enables a future in which OPE can provide rigorous uncertainty estimates for high-stakes domains.", "date": "2025-07-26", "categories": []} {"id": "2602.13848", "title": "Testing For Distribution Shifts with Conditional Conformal Test Martingales", "abstract": "We propose a sequential test for detecting arbitrary distribution shifts that allows conformal test martingales (CTMs) to work under a fixed, reference-conditional setting. Existing CTM detectors construct test martingales by continually growing a reference set with each incoming sample, using it to assess how atypical the new sample is relative to past observations. While this design yields anytime-valid type-I error control, it suffers from test-time contamination: after a change, post-shift observations enter the reference set and dilute the evidence for distribution shift, increasing detection delay and reducing power.\n In contrast, our method avoids contamination by design by comparing each new sample to a fixed null reference dataset. Our main technical contribution is a robust martingale construction that remains valid conditional on the null reference data, achieved by explicitly accounting for the estimation error in the reference distribution induced by the finite reference set. This yields anytime-valid type-I error control together with guarantees of asymptotic power one and bounded expected detection delay. Empirically, our method detects shifts faster than standard CTMs, providing a powerful and reliable distribution-shift detector.", "date": "2026-02-14", "categories": []} {"id": "2605.00432", "title": "Optimal Spatio-Temporal Decoupling for Bayesian Conformal Prediction", "abstract": "Online conformal prediction must balance fast adaptation to distribution shift against stable coverage: feedback-driven methods react quickly but become volatile, while strongly discounted Bayesian methods lag and inflate intervals at tight coverage. We introduce \\textbf{State-Adaptive Bayesian Conformal Prediction (SA-BCP)}, which forms the predictive quantile as a gated convex combination of long-term temporal inertia and local spatial evidence from a kernel density estimate, controlled by a single interpretable evidence threshold $K$. We establish three results: (i) asymptotic marginal validity of the resulting intervals; (ii) a closed-form expression for the MSE-optimal threshold, $K^*_{\\mathrm{MSE}}=\\alpha(1-\\alpha)/M^{\\mathcal{T}}$, trading the coverage-indicator (Bernoulli) variance against the temporal structural bias $M^{\\mathcal{T}}$; and (iii) a rolling-origin procedure for selecting $K$ online -- consistent under stationarity, with $O(\\sqrt{T\\log N})$ regret against the best fixed $K$ and, for a segmented variant, a sublinear dynamic-regret bound under bounded drift. Across four financial-volatility and weather datasets, three target coverage levels, and eight baselines (including the strongest recent conditional-quantile methods, SPCI and KOWCPI), SA-BCP attains at-or-above-nominal coverage in most settings while producing substantially sharper intervals -- up to roughly $3\\times$ lower Winkler score than discounted Bayesian CP at the tightest coverage -- and a coverage-matched audit confirms these efficiency gains are not an artifact of under-coverage. We disclose one principal limitation: a volatility-specialized conformal-GARCH competitor remains more efficient on its home volatility-base series, though it does not transfer across domains.", "date": "2026-05-01", "categories": []} {"id": "2606.12501", "title": "Policy-driven Conformal Prediction for Trustworthy QoT Estimation", "abstract": "We propose Conformal QoT, a policy-driven framework that combines statistically guaranteed QoT estimation with operational decision policies, enabling reliable lightpath-feasibility predictions under domain shift and improving accuracy from 92\\% to 99.6\\% on open datasets.", "date": "2026-06-10", "categories": []} {"id": "2410.24145", "title": "Projected random forests and conformal prediction of circular data", "abstract": "We apply conformal prediction techniques to regression problems with circular responses, producing prediction sets with adaptive arc length and finite-sample coverage guarantees for any circular predictive model under the assumption of data exchangeability. Leveraging the high performance of existing predictive models designed for linear responses, we analyze a general projection procedure that converts any linear-response regression model into one suitable for circular responses. When random forests are used as base models in this projection procedure, we leverage the random forest out-of-bag mechanism to eliminate the need for a separate calibration sample in the construction of prediction sets. On synthetic and real datasets, the resulting projected random forest model produces more efficient out-of-bag conformal prediction sets, with shorter median arc length, than the split conformal prediction sets generated by two existing alternative models.", "date": "2024-10-31", "categories": []} {"id": "2411.12193", "title": "Hierarchical Probabilistic Conformal Prediction for Distributed Energy Resources Adoption", "abstract": "The rapid growth of distributed energy resources (DERs) presents both opportunities and operational challenges for electric grid management. Accurately predicting DER adoption is critical for proactive infrastructure planning, but the inherent uncertainty and spatial disparity of DER growth complicate traditional forecasting approaches. Moreover, the hierarchical structure of distribution grids demands that predictions satisfy statistical guarantees at both the circuit and substation levels, a non-trivial requirement for reliable decision-making. In this paper, we propose a novel uncertainty quantification framework for DER adoption predictions that ensures validity across hierarchical grid structures. Leveraging a multivariate Hawkes process to model DER adoption dynamics and a tailored split conformal prediction algorithm, we introduce a new nonconformity score that preserves statistical guarantees under aggregation while maintaining prediction efficiency. We establish theoretical validity under mild conditions and demonstrate through empirical evaluation on customer-level solar panel installation data from Indianapolis, Indiana that our method consistently outperforms existing baselines in both predictive accuracy and uncertainty calibration.", "date": "2024-11-19", "categories": []} {"id": "2602.14913", "title": "Coverage Guarantees for Pseudo-Calibrated Conformal Prediction under Distribution Shift", "abstract": "Conformal prediction (CP) offers distribution-free marginal coverage guarantees under an exchangeability assumption, but these guarantees can fail if the data distribution shifts. We analyze the use of pseudo-calibration as a tool to counter this performance loss under a bounded label-conditional covariate shift model. Using tools from domain adaptation, we derive a lower bound on target coverage in terms of the source-domain loss of the classifier and a Wasserstein measure of the shift. Using this result, we provide a method to design pseudo-calibrated sets that inflate the conformal threshold by a slack parameter to keep target coverage above a prescribed level. Finally, we propose a source-tuned pseudo-calibration algorithm that interpolates between hard pseudo-labels and randomized labels as a function of classifier uncertainty. Numerical experiments show that our bounds qualitatively track pseudo-calibration behavior and that the source-tuned scheme mitigates coverage degradation under distribution shift while maintaining nontrivial prediction set sizes.", "date": "2026-02-16", "categories": []} {"id": "2606.05551", "title": "Conformal Risk-Averse Decision Making with Action Conditional Guarantee", "abstract": "Reliable decision making pipelines powered by machine learning models require uncertainty quantification (UQ) methods that come with explicit safety guarantees. Conformal prediction provides such UQ by wrapping ML predictions into prediction sets, and recent work by Kiyani et al. (2025b) established that these sets can be translated into optimal risk-averse decision policies -- yet only inheriting marginal safety guarantees. We generalize and strengthen their results by (i) introducing action-conditional conformal prediction, which yields safety guarantees conditioned explicitly on each action taken by the decision maker, (ii) showing that action-conditional prediction sets serve as a proxy for the feasible decision space for risk-averse decision makers aiming to optimize action-conditional value-at-risk, and (iii) proposing a principled finite-sample algorithm based on pinball-loss minimization, connecting the framework of Gibbs et al. (2025) to action-conditional guarantees. Experiments on two real-world datasets confirm that our approach significantly improves action-conditional performance over conformal baselines.", "date": "2026-06-04", "categories": []} {"id": "2606.11865", "title": "Conformal Bayes under Label Shift: Post-Hoc Calibration vs. In-Training Adaptation", "abstract": "Conformal Bayes combines Bayesian posterior predictives with conformal calibration to produce prediction sets that are both statistically valid and geometrically efficient. We study conformal Bayes under label shift from a unified perspective, identifying two complementary approaches that restore nominal target-domain coverage through importance-weighted conformal calibration but operate through independent mechanisms. \\emph{Post-hoc calibration} tilts the posterior predictive toward the target domain and corrects the conformal threshold via an importance-weighted quantile, leaving the parameter posterior unchanged. \\emph{In-training adaptation} tilts the parameter posterior itself to the target domain, producing a corrected predictive whose highest predictive density region serves as the highest predictive density (HPD) based prediction set under the fitted target predictive; efficiency is model-dependent and does not imply finite-sample conditional optimality. Two controlled experiments show that in an unbiased training regime both strategies achieve valid coverage equally, while in a lead-optimization regime in-training adaptation acts as a debiasing operator, reducing interval width at unchanged coverage.", "date": "2026-06-10", "categories": []} {"id": "2606.11949", "title": "Online Shift Detection and Conformal Adaptation for Deployed Safety Classifiers", "abstract": "We present an online monitoring system for distributional shift in deployed safety classifiers, using calibrated sequential statistics to detect when a classifier has moved out of distribution. Upon detection, a conformal abstention layer adapts decision thresholds to recover a target error rate epsilon=0.1. In a pre-registered factorial evaluation (4 classifiers x 5 shift conditions x 20 seeds x 2 window sizes, 800 cells), the system achieves 86.6% valid detection (693/800, 95% CI [84.1%, 88.8%]) with mean latency of 39.5 steps. Detection holds across three ground-truth regimes: synthetic onset (86.6%), real temporal jailbreaks (85%, 17/20), and GCG adversarial attacks. Weighted conformal prediction recovers up to 39 pp of lost coverage for DeBERTa (ESS=46/300) but collapses for all other classifiers (ESS~300): logistic density ratio estimation achieves perfect source/target separability in high-dimensional embedding spaces, clipping all importance weights to the floor. DeBERTa shows a gradient from effective correction (paraphrase, ESS=46) to near-total collapse (adversarial suffix, ESS=206). PCA to 32 dimensions breaks the collapse, recovering 33 pp for Llama Guard and 21 pp for ShieldGemma. Variance decomposition reveals classifier (eta^2=0.243), shift type (eta^2=0.237), and their interaction (eta^2=0.185) all contribute substantially to detection latency variance (all p<0.001), indicating per-classifier monitoring profiles are necessary.", "date": "2026-06-10", "categories": []} {"id": "2508.13362", "title": "Optimization-based Online Conformal Prediction for Multi-step Forecasting", "abstract": "Conformal prediction (CP) is well-suited for uncertainty quantification in time series forecasting due to its distribution-free coverage guarantees. However, existing multi-step methods often struggle to balance coverage validity with efficiency: they either calibrate horizons independently, ignoring temporal correlations, or enforce strict simultaneous coverage, resulting in overly conservative intervals. In this work, we propose O2CP: Optimization-based Online Conformal Prediction, a unified framework for online conformal prediction that explicitly models multi-step error dependencies without sacrificing long-term marginal coverage guarantees. We first prove that standard online conformal updates maintain validity as long as calibration parameters remain within a defined \"safe\" region. Leveraging this theoretical insight, we introduce a two-layer architecture: an outer layer that defines admissible parameter sets to ensure validity, and an inner layer that performs constrained optimization to model joint error distributions and minimize horizon-wide objectives. To make this computationally feasible, we develop a lightweight sampling strategy that estimates joint distributions without requiring large calibration sets. Extensive experiments on real-world datasets, including autonomous driving, climate forecasting, and public health, demonstrate that O2CP consistently outperforms state-of-the-art baselines, achieving target coverage with significantly sharper prediction intervals and reduced regret over long horizons.", "date": "2025-08-18", "categories": []} {"id": "2605.30292", "title": "Leave a Window Out: Modifying the Jackknife for Predictive Inference in Time Series", "abstract": "Conformal prediction methods enjoy strong theoretical and empirical predictive inference performance, provided the data is exchangeable and is treated symmetrically during training. However, these assumptions are impractical in many settings, such as time series, where temporal dependence violates exchangeability and it is preferable to use predictors that leverage dependence by treating data asymmetrically. Recent work shows that split conformal prediction is robust to these issues, but sample splitting can reduce accuracy, motivating the study of methods that do not rely on data splitting in the time series setting.\n In this work, we show that the vanilla leave-one-out jackknife can suffer arbitrary loss of coverage even in canonical time series models with mild temporal dependence. As a remedy, we propose a modification tailored to such settings, which we term the leave-a-window-out (LWO) method, and show that it can achieve valid coverage provided that the model-fitting procedure satisfies mild stability properties. Our proofs are based on quantifying the degree to which the data departs from cyclic exchangeability, which we introduce new coefficients to measure. Experiments on time series demonstrate that our method often enjoys valid coverage when the vanilla jackknife fails to cover, while producing much narrower intervals than split conformal prediction.", "date": "2026-05-28", "categories": []} {"id": "2606.09860", "title": "Conformal Risk Prediction for Non-Alcoholic Fatty Liver Disease Using Gradient Boosting with Distribution-Free Coverages", "abstract": "Non-alcoholic fatty liver disease (NAFLD) affects roughly 25% of global adults, posing substantial hepatic and cardiovascular risks. Yet, population-level screening tools remain inadequate. We present Method, a machine-learning framework for NAFLD risk prediction coupling gradient-boosted decision trees with conformal prediction to yield calibrated, distribution-free coverage guarantees on individual risk estimates. It integrates a mutual-information-based stability selection procedure to identify a compact, clinically interpretable feature subset via bootstrap resampling, constructing prediction sets whose marginal coverage provably exceeds a user-specified confidence level. We evaluated Method on a multicenter cohort from Guangzhou, China (primary n=2,187; external validation n=412) using 78 candidate features across demographics, metabolic biomarkers, and lifestyle factors. Method achieves an AUROC of 0.912 internally and 0.891 externally, outperforming deep neural networks, TabNet, support vector machines, and logistic regression. Conformal prediction sets achieve 91.3% empirical coverage at the 90% nominal level. A three-tier risk stratification derived from these scores separates the population into distinct groups, with the high-risk subgroup showing a 12-month progression rate 4.7 times that of the low-risk tier. The selected features -- notably waist circumference, ALT, GGT, triglycerides, fasting glucose, and BMI -- align with established metabolic risk factors, providing biological plausibility.", "date": "2026-05-31", "categories": []} {"id": "2606.10187", "title": "Decision-Calibrated Conformal Uncertainty for Pacing Decisions in Streaming Advertising", "abstract": "We develop a decision-calibrated conformal framework for pacing decisions in streaming advertising. Pacing depends on uncertain future inventory, demand pressure, incremental response, and member-experience load. Instead of calibrating a generic forecast residual, the framework measures forecast error by its largest impact on the policies that could actually be deployed. The main theorem shows that the proposed score is the smallest valid uncertainty measure that uniformly protects all deployable pacing policies. Geometrically, it is the support function of the signed policy sensitivity set. Split conformal calibration gives finite-sample coverage for this score. A high-dimensional separation theorem shows that traditional residual calibration can be arbitrarily more conservative by paying for nuisance inventory dimensions, and a robust pacing result combines inventory, response, and experience uncertainty. On public-data-calibrated pacing replays built from Criteo Uplift and KuaiRand datasets, traditional conformal pacing remains unresolved with high residual radii of 7236.7 on Criteo and 4629.4 on KuaiRand. With the proposed decision calibration approach, the uncertainty radii are reduced to 18.4 and 278.6 respectively, with separate margins for value, delivery, budget, and member load. On Criteo, the proposed method certifies a less aggressive pacing policy than the point-forecast baseline, and reduces held-out any-violation rate from 16.7% to 3.3%, with zero budget and member-load violations. On KuaiRand, the choice remains unresolved. In a nutshell, the paper establishes that forecasts, response estimates, and member-experience models should be judged by whether they shrink the uncertainty that the pacing decision uses, as this leads to confident decisions that are not overly conservative.", "date": "2026-06-08", "categories": []} {"id": "2606.10734", "title": "SPACR: Single-Pass Adaptive Training of Uncertainty-Aware Conformal Regressors", "abstract": "Conformal Prediction (CP) provides robust uncertainty guarantees for predictive models, but is typically applied post hoc, which misaligns model training with the conformal goal of producing efficient (i.e, narrow) intervals. We propose SPACR (Single-Pass Adaptive Conformal Regressor), a novel method for directly training uncertainty-aware regressors within a differentiable loss. SPACR jointly optimizes efficiency and validity without batch-splitting or a predefined confidence levels during training. As a result, a single SPACR model yields valid prediction intervals at multiple confidence levels during inference, avoiding the costly retraining required by methods like DOICR. Experiments on diverse datasets show that SPACR consistently gives tighter intervals and better coverage-efficiency trade-offs compared to standard CP and DOICR, while significantly reducing computational costs.", "date": "2026-06-09", "categories": []} {"id": "2606.11044", "title": "Generalized Conformal Predictive Systems Under Distributional Shifts", "abstract": "Conformal predictive systems (CPS) output calibrated bands of CDFs under exchangeability. We extend generalized CPS to non-exchangeable settings by encoding distributional shifts through observation-specific permutation weights. This yields shift-aware predictive systems that remain valid whenever the test point is, conditionally on the unordered sample, a weighted draw from the observed atoms. Since such weights are typically estimated, we introduce weight-uncertainty boxes and construct robust CPS envelopes with finite-sample or asymptotic confidence guarantees. We derive efficient computation for conformity-measure CPS, conformal binning, and conformal isotonic distributional regression. Experiments under covariate shift and feedback-driven biomolecular design show calibrated predictive bands that widen under stronger shifts and tighten as sample size increases.", "date": "2026-06-09", "categories": []} {"id": "2507.20975", "title": "Locally Adaptive Conformal Inference for Operator Models", "abstract": "Operator models are regression algorithms between Banach spaces of functions. They have become an increasingly critical tool for spatiotemporal forecasting and physics emulation, especially in high-stakes scenarios where robust, calibrated uncertainty quantification is required. We introduce Local Sliced Conformal Inference (LSCI), a distribution-free framework for generating function-valued, locally adaptive prediction sets for operator models. We prove finite-sample validity and derive a data-dependent upper bound on the coverage gap under local exchangeability. On synthetic Gaussian-process tasks and real applications (air quality monitoring, energy demand forecasting, and weather prediction), LSCI yields tighter sets with stronger adaptivity compared to conformal baselines. We also empirically demonstrate robustness against biased predictions and certain out-of-distribution noise regimes.", "date": "2025-07-28", "categories": []} {"id": "2602.16015", "title": "Geometry-Aware Uncertainty Quantification via Conformal Prediction on Manifolds", "abstract": "Conformal prediction gives finite-sample coverage guarantees for regression, but most standard constructions are designed for Euclidean output spaces. When the response lies on a Riemannian manifold, Euclidean residuals and coordinate-based regions can ignore the geometry that defines meaningful error. We propose adaptive geodesic conformal prediction, a simple framework that builds nonconformity scores from geodesic distances and normalizes them with a cross-validated estimate of local prediction difficulty. On the sphere, this produces geodesic caps whose area is independent of position, while their radii still adapt to heteroscedastic noise. In both a synthetic sphere experiment and an IGRF-14 geomagnetic field forecasting task, the adaptive method preserves valid marginal coverage, reduces variation in conditional coverage, and improves worst-case coverage relative to non-adaptive and coordinate-based baselines.", "date": "2026-02-17", "categories": []} {"id": "2606.00419", "title": "Parameter-Free and Group Conditional Online Conformal Prediction", "abstract": "Uncertainty quantification (UQ) is critical for the deployment of machine learning predictors in real-world scenarios where the data distribution may shift over time (i.e., data may not be exchangeable). Online conformal prediction (OCP) methods address this issue at the expense of either (i) group-wise error control or (ii) learning-rate independent implementation. Group-conditional coverage is essential for fairness across different collections of data points and for providing finer UQ guarantees. Parameter-free optimization is crucial for robustness to adversarial and unknown data shifts. We propose a parameter-free algorithm for group-conditional OCP and demonstrate that it achieves the best group-conditional coverage guarantees. We evaluate our algorithm on synthetic and real-world data, demonstrating that our method not only improves the reliability of existing parameter-free OCP methods but also provides prediction intervals that are comparable in size to well-tuned group-conditional approaches. By unifying group-conditional coverage with parameter-free online algorithms, our work lays a foundation for fair and robust uncertainty quantification in shifting environments.", "date": "2026-05-29", "categories": []} {"id": "2606.07771", "title": "Beyond Point Estimates: Benchmarking Uncertainty Quantification Methods on the AION-1 Astronomical Foundation Model", "abstract": "Foundation models for astronomical surveys offer powerful learned representations that can be transferred to downstream regression tasks such as galaxy property estimation. However, point predictions alone are insufficient for scientific inference; reliable uncertainty quantification (UQ) is essential. We compare seven UQ methods on galaxy property regression using frozen AION-1 foundation-model embeddings, predicting redshift, stellar mass, stellar-population age, gas-phase metallicity, and specific star-formation rate, from Legacy Survey photometry/imaging and DESI spectra, with PROVABGS-derived labels. Distribution-free conformal methods achieve marginal coverage within $\\sim$1\\,pp of the nominal 90\\% across all properties, while non-conformal baselines (Deep Ensembles, MC~Dropout) fail to calibrate reliably. Among conformal approaches, Conformalized Quantile Regression (CQR) delivers the best coverage in the bin with the poorest model predictions. More importantly, only the Locally Valid and Discriminative (LVD) framework -- particularly when operating on AION-1 embeddings -- also provides finite-sample \\emph{local validity}, producing intervals that adapt to each galaxy's local prediction difficulty rather than relying on marginal guarantees alone. These results establish conformal prediction, and LVD in particular, as the preferred UQ framework for uncertainty-aware inference on foundation-model embeddings in astrophysics.", "date": "2026-06-05", "categories": []} {"id": "2606.08517", "title": "A Joint Finite-Sample Certificate for Adaptive Selective Conformal Risk Control", "abstract": "Selective predictors answer on confident inputs and abstain elsewhere; deploying one safely needs a single finite-sample certificate that simultaneously upper-bounds the selected risk, lower-bounds the acceptance probability $\\pacc$ above a floor $\\pmin$, and lower-bounds the deployment utility. This certificate must be valid under adaptive threshold selection from a finite grid of $m$ pairs on $\\ncert$ samples. We give such a certificate for bounded, possibly non-monotone losses by treating the selected risk directly as a ratio rather than through a Hoeffding-style range bound. The construction couples three confidence bounds: a variance-adaptive empirical-Bernstein bound on the ratio risk, a Clopper--Pearson bound on acceptance, and a two-sided closeness bound on utility. Together they lower-bound the certified policy's utility absolutely and to within $2\\gammau$ of the best over the \\emph{certified set}, both non-vacuous whenever feasible; a regime-scoped third leg matches an external oracle, informative only where the risk margin $\\gammar < \\alpha$ and vacuous at the headline operating points. Relative to the range-only Hoeffding-ratio construction this sharpens the acceptance-floor dependence from $1/\\pmin$ to $1/\\sqrt{\\pmin}$, and a closed-form corollary identifies a per-pair regime in which our risk bound dominates a Hoeffding conformal risk control (Hoeffding--CRC) selective bound. Empirically, on ImageNet (three ResNets) and COCO val 2017 panoptic, the certificate opens a $+22$ pp certified-acceptance frontier over Hoeffding--CRC and is ${\\approx}10{\\times}$ tighter than a non-vacuous matched-valid baseline; these gains are regime-scoped, not universal, and absent on ADE20K. The certifier runs in $O(\\ncert m)$ time.", "date": "2026-06-07", "categories": []} {"id": "2606.08654", "title": "Operator learning for the 2D incompressible Navier-Stokes equations: a conformal prediction approach in the data-scarce regime", "abstract": "In this paper, we propose a perturbation-based conformal prediction framework for uncertainty quantification in operator learning, with a focus on the 2D Navier--Stokes equations. While neural operators provide fast surrogates for expensive PDE solvers, they do not by themselves provide calibrated uncertainty for spatiotemporal field predictions. Our approach wraps a trained Fourier Neural Operator (FNO) with split conformal prediction and constructs the local uncertainty scale by comparing the predictions of two operators trained on nearly identical datasets: one on the original labels and one on labels perturbed by small Gaussian noise. We consider this procedure in the data-scarce regime, where the total label budget is fixed and methods that require a separate uncertainty network must divide training data between multiple models. On the 2D Navier--Stokes benchmark, the perturbation-based method produces substantially narrower conformal bands than existing methods under matched total data budgets while maintaining the target simultaneous coverage. These results suggest that perturbation sensitivity is a practical and sample-efficient uncertainty proxy for conformalized neural operators.", "date": "2026-06-07", "categories": []} {"id": "2606.08831", "title": "Inference-Time Conformal Reasoning with Valid Factuality Control for Large Language Models", "abstract": "Large language models (LLMs) increasingly perform multi-step reasoning, where intermediate claims form implicit directed acyclic graphs whose node correctness is structurally conditioned on their ancestors. This makes factuality uncertainty structural, rather than a trivial accumulation of node-wise errors, and necessitates inference-time uncertainty quantification over the reasoning structure. While conformal prediction (CP) offers flexible user-specified factuality control, existing work remains post-hoc and cannot intervene during generation. To fill the gap between CP's flexibility and its post-hoc limitation, we propose an \\emph{Inference-Time Conformal Reasoning (ITCR)} framework that integrates CP directly into reasoning graph generation. ITCR learns a structure-level factuality uncertainty function that aggregates claim-level factuality signals over reasoning graphs without complex modeling assumptions. We then design the non-conformity score based on graph-level factuality uncertainty and calibrate the conformal threshold to decide when to stop generation. We theoretically show such generation is nested, yielding valid coverage guarantees for factuality control. Experiments over multiple datasets and coverage objectives demonstrate empirically valid coverage. In downstream reasoning tasks, inference-time calibrated graphs yield more accurate generation than post-hoc pruned graphs.", "date": "2026-06-07", "categories": []} {"id": "2606.09473", "title": "Report the Floor: A Training-Free Conformal Interval Is a Mandatory Baseline for Probabilistic Time-Series Forecasting", "abstract": "Probabilistic forecasters are increasingly learned, yet the baselines they are compared against are often weak or omitted. We show that the simplest possible conformal interval - a last-value point forecast wrapped in a finite-sample split-conformal residual quantile, with no parameters and no training - is a far stronger baseline than its near-total absence from recent learned-forecasting and conformal-time-series comparisons would suggest. In one-step-ahead online forecasting across 2,217 real series from nine public sources (Monash, LOTSA, the LTSF traffic/electricity/weather suites, METR-LA, BOOM, nips/probts), this ConformalNaive interval decisively beats the naive value-quantile baselines, the entire NPTS family (NPTS 73%, SeasonalNPTS 64% of series), and the published Conformal Seasonal Pools (CSP) method (71% of series, bootstrap 95% CI [69,73], paired Wilcoxon p approx 7.6e-135); it is on par with the simpler learned conformal predictors (RCI, quantile regression; median relative Winkler within 2%) and is beaten only by the adaptive-online and ensemble methods (SPCI, ACI, AgACI), which track distribution shift and lead by 9-33% relative Winkler. It is also better calibrated than a trained neural forecaster: on the six datasets that introduced DeepNPTS, the trivial floors cover the truth 84-85% of the time at a nominal 95%, versus DeepNPTS's 66%. At multi-step seasonal horizons the picture inverts: the random-walk floor is the weakest method and the seasonal pool (CSP) wins - a boundary we map. Finally we give ConformalNaive+, a one-line, training-free, horizon-adaptive selector that attains the better of two complementary floors at every horizon with restored coverage. We argue the matching conformal naive floor must be a mandatory baseline whenever a learned probabilistic forecaster claims gains.", "date": "2026-06-08", "categories": []} {"id": "2511.19359", "title": "Enhancing Conformal Prediction via Class Similarity", "abstract": "Conformal Prediction (CP) has emerged as a powerful statistical framework for high-stakes classification applications. Instead of predicting a single class, CP generates a prediction set, guaranteed to include the true label with a pre-specified probability. The performance of different CP methods is typically assessed by their average prediction set size. In setups where the classes can be partitioned into semantic groups, e.g., diseases that require similar treatment, users can benefit from prediction sets that are not only small on average, but also contain a small number of semantically different groups. This paper begins by addressing this problem and ultimately offers a widely applicable tool for boosting any CP method on any dataset. First, given a class partition, we propose augmenting the CP score function with a term that penalizes predictions with out-of-group errors. We theoretically analyze this strategy and prove its advantages for group-related metrics. Surprisingly, we show mathematically that, for common class partitions, it can also reduce the average set size of any CP score function. Our analysis reveals the class-similarity factors behind this improvement and motivates a variant that can further reduce prediction set size by leveraging the model's embeddings, without requiring any human semantic partition. Finally, we present an extensive empirical study, encompassing prominent CP methods, multiple models, and several datasets, which demonstrates that our class-similarity-based approach consistently enhances CP methods.", "date": "2025-11-24", "categories": []} {"id": "2606.06391", "title": "Conformal Risk Sharing: Certified Cost Allocation with Participation Guarantees", "abstract": "Sharing the financial impact of rare adverse events across a group can soften extreme individual burdens, but any participant made worse off by the arrangement has reason to leave. A credible mechanism must therefore provide each agent with a trustworthy cap on their future obligation and should be deployed only if the aggregate harm across participants is bounded. We formalise this as the Certified Allocation Problem: from finite data and without distributional assumptions, find a redistribution rule, produce obligation caps for every participant, and verify that no participant is made materially worse off. We propose Conformal Risk Sharing, which solves this problem by pairing an interpretable sharing policy with split conformal calibration. The sharing intensity is tuned on training data, while held-out calibration data produces distribution-free per-agent guarantees (valid under exchangeability). Experiments on synthetic and real-world data, including precipitation and energy-cooperative data, confirm that the framework can substantially reduce extreme obligations for high-risk agents while controlling harm to others.", "date": "2026-06-04", "categories": []} {"id": "2605.20468", "title": "CASCADE Conformal Prediction: Uncertainty-Adaptive Prediction Intervals for Two-Stage Clinical Decision Support", "abstract": "Effective medication management in Parkinson's Disease (PD) is challenging due to heterogeneous disease progression, variable patient response, and medication side effects. While AI models can forecast levodopa equivalent daily dose (LEDD) as a measure of medication needs, standard uncertainty quantification often fails to communicate the reliability of these predictions, treating high and low confidence clinical decisions identically. We introduce CASCADE (Calibrated Adaptive Scaling via Conformal And Distributional Estimation), a novel conformal prediction framework that propagates epistemic uncertainty from a screening classifier to adapt downstream predictions. Unlike standard conformal methods that rely on auxiliary residual regression, we leverage epistemic uncertainty from a primary classification task (identifying whether a medication change is needed) to dynamically scale the prediction intervals of a secondary regression task (predicting how much change). By mapping Venn-Abers multi-probabilistic uncertainty directly to non-conformity scores, our framework achieves continuous risk adaptation. We demonstrate that this ``cascade effect'' produces highly efficient intervals for confident patients (38.9% narrower than standard conformal baselines) while automatically expanding intervals to ensure robust coverage for uncertain cases, bridging the gap between discrete clinical decision-making and continuous dose forecasting in PD.", "date": "2026-05-19", "categories": []} {"id": "2505.11785", "title": "Improving Coverage in Combined Prediction Sets with Weighted p-values", "abstract": "Conformal prediction quantifies the uncertainty of machine learning models by augmenting point predictions with valid prediction sets. For complex scenarios involving multiple trials, models, or data sources, conformal prediction sets can be aggregated to create a prediction set that captures the overall uncertainty, often improving precision. However, aggregating multiple prediction sets with individual $1-\\alpha$ coverage inevitably weakens the overall guarantee, typically resulting in $1-2\\alpha$ worst-case coverage. In this work, we propose a framework for the weighted aggregation of prediction sets, where weights are assigned to each prediction set based on their contribution. Our framework offers flexible control over how the sets are aggregated, achieving tighter coverage bounds that interpolate between the $1-2\\alpha$ guarantee of the combined models and the $1-\\alpha$ guarantee of an individual model depending on the distribution of weights. Importantly, our framework generalizes to data-dependent weights, as we derive a procedure for weighted aggregation that maintains finite-sample validity even when the weights depend on the data. This extension makes our framework broadly applicable to settings where weights are learned, such as mixture-of-experts (MoE), and we demonstrate through experiments in the MoE setting that our methods achieve adaptive coverage.", "date": "2025-05-17", "categories": []} {"id": "2606.02911", "title": "The Ghost Annotator: a Framework to Explore Human Label Variation in Content Moderation through Conformal Prediction", "abstract": "Current research primarily focuses on model performance, while comparatively less attention has been devoted to uncertainty estimation, particularly in settings where LLMs are increasingly used to generate annotated data. We introduce a framework combining conformal prediction with Collaborative Filtering-style annotators' representation to model LLM behavior in relation to human annotators and to analyze patterns of agreement and disagreement. Using Non-Conformity Scores, we introduce the Ghost Prediction metric and the Ghost Annotator representation to quantify cases in which model predictions diverge from all available human annotations. We compute cosine similarity measures to explore differences in model behavior across sociodemographic axes. We evaluated four LLMs of different size and families across four content moderation datasets. Our finding shows that while we find that all models uncertainty increases with annotator disagreement, larger models tend to be more confident in the classification of texts that are not aligned with any human annotation. Finally, the Ghost Annotator framework reveals a consistent and robust pattern of demographic misalignment, suggesting a structural bias likely rooted in pretraining corpora.", "date": "2026-06-01", "categories": []} {"id": "2606.03600", "title": "Set-Preserving Calibration from Conformal P-Values to E-Values", "abstract": "Standard conformal prediction (CP) procedures are typically formulated in terms of p-values, but reliance on p-values alone limits flexibility, for example, when combining dependent evidence across models or data splits. Recent work has explored e-value formulations for conformal inference, yet a direct connection between p- and e-value formulations in CP has been missing, especially regarding their statistical efficiency. We first identify limitations of classical p-to-e calibrators in the CP setting, showing that they are not set-preserving and can lead to overly conservative prediction sets. To address this, we propose a novel P2E calibrator that converts conformal p-values into e-values without altering the prediction set induced by the original conformal p-value. We establish both theoretically and empirically that our calibrator can yield significant efficiency gains over existing p-to-e calibrators. This e-value formulation enables principled use of recent advances in e-value merging and randomization, where we demonstrate its impact in two applications: cross-conformal prediction (CCP), whose variants typically provide only approximate $1-2\\alpha$ coverage, and conformal aggregation (CA). In both cases, our e-value-based methods satisfy the desired $1-\\alpha$ coverage guarantee while improving efficiency over standard baselines. More broadly, our approach expands the flexibility of CP and opens new directions for efficient, distribution-free uncertainty quantification.", "date": "2026-06-02", "categories": []} {"id": "2606.03731", "title": "Conformal Language Modeling via Posterior Sampling", "abstract": "Large Language Models remain plagued by hallucinations. Recent work has sought to tame their prevalence using statistical techniques based on conformal prediction, with both theoretical and empirical success. However, these methods operate in a post-hoc fashion, treating the sampling procedure itself as atomic and then surgically altering samples to remove hallucinated claims. This disconnect between filtering and generation can result in samples that are incoherent, inconsistent, or simply unlikely under the model itself. Moreover, post-hoc surgery is unable to shift probability mass towards more useful and helpful responses. To address these issues, we propose to instead sample from approximations to an LLM posterior, where the conditioning event corresponds to a calibrated, high-scoring region. We develop a calibration procedure tailored to the setting of conditional sequential generation that effectively identifies this region and achieves target risk control. Empirically, we apply our method to case studies focused on open-ended biography generation and mathematical problem solving; compared to prior work, we obtain the same statistical guarantees, with higher downstream utility.", "date": "2026-06-02", "categories": []} {"id": "2509.04631", "title": "Fundamental bounds on efficiency-confidence trade-off for transductive conformal prediction", "abstract": "Transductive conformal prediction addresses the simultaneous prediction for multiple data points. Given a desired confidence level, the objective is to construct a prediction set that includes the true outcomes with the prescribed confidence. We demonstrate a fundamental trade-off between confidence and efficiency in transductive methods, where efficiency is measured by the size of the prediction sets. Specifically, we derive a strict finite-sample bound showing that any non-trivial confidence level leads to exponential growth in prediction set size for data with inherent uncertainty. The exponent scales linearly with the number of samples and is proportional to the conditional entropy of the data. Additionally, the bound includes a second-order term, dispersion, defined as the variance of the log conditional probability distribution. We show that the transductive methods based on the approximate conditional distribution can approach this bound. Inspired by this setup, we introduce a practical transductive prediction algorithm that surpasses Bonferroni methods.", "date": "2025-09-04", "categories": []} {"id": "2509.22907", "title": "FedCF: Fair Federated Conformal Prediction", "abstract": "Conformal Prediction (CP) is a widely used technique for quantifying uncertainty in machine learning models. In its standard form, CP offers probabilistic guarantees on the coverage of the true label, but it is agnostic to sensitive attributes in the dataset. Several recent works have sought to incorporate fairness into CP by ensuring conditional coverage guarantees across different subgroups. One such method is Conformal Fairness (CF). In this work, we extend the CF framework to the Federated Learning setting and discuss how we can audit a federated model for fairness by analyzing the fairness-related gaps for different demographic groups. We empirically validate our framework by conducting experiments on several datasets spanning multiple domains, fully leveraging the exchangeability assumption.", "date": "2025-09-26", "categories": []} {"id": "2510.05566", "title": "Domain-Shift-Aware Conformal Prediction for Large Language Models", "abstract": "Large language models have achieved impressive performance across diverse tasks. However, their tendency to produce overconfident and factually incorrect outputs, known as hallucinations, poses risks in real-world applications. Conformal prediction provides finite-sample, distribution-free coverage guarantees, but standard conformal prediction breaks down under domain shift, often leading to under-coverage and unreliable prediction sets. We propose a new framework called Domain-Shift-Aware Conformal Prediction (DS-CP). Our framework adapts conformal prediction to large language models under domain shift, by systematically reweighting calibration samples based on their proximity to the test prompt, thereby preserving validity while enhancing adaptivity. Our theoretical analysis and experiments on the MMLU benchmark demonstrate that the proposed method delivers more reliable coverage than standard conformal prediction, especially under substantial distribution shifts, while maintaining efficiency. This provides a practical step toward trustworthy uncertainty quantification for large language models in real-world deployment.", "date": "2025-10-07", "categories": []} {"id": "2602.16794", "title": "Beyond Procedure: Substantive Fairness in Conformal Prediction", "abstract": "Conformal prediction (CP) offers distribution-free uncertainty quantification for machine learning models, yet its interplay with fairness in downstream decision-making remains underexplored. Moving beyond CP as a standalone operation (procedural fairness), we analyze the holistic decision-making pipeline to evaluate substantive fairness-the equity of downstream outcomes. Theoretically, we derive an upper bound that decomposes prediction-set size disparity into interpretable components, clarifying how label-clustered CP helps control method-driven contributions to unfairness. To facilitate scalable empirical analysis, we introduce an LLM-in-the-loop evaluator that approximates human assessment of substantive fairness across diverse modalities. Our experiments show that label-clustered CP often provides a favorable balance between utility and substantive fairness, while reducing set-size disparities in line with our theory. Finally, we empirically show that equalized set sizes, rather than coverage, strongly correlate with improved substantive fairness, enabling practitioners to design more fair CP systems. Our code is available at https://github.com/layer6ai-labs/llm-in-the-loop-conformal-fairness.", "date": "2026-02-18", "categories": []} {"id": "2606.00320", "title": "Adversarially Robust Control of Conditional Value-at-Risk via Rockafellar-Uryasev Conformal Inference", "abstract": "We present an online, distribution-free framework for controlling the Conditional Value-at-Risk (CVaR), extending conformal tail risk control to non-stationary and adversarial environments. Unlike classical risk control methods, which rely on stationarity or linearity of expectation, our approach provides provable safety guarantees for a nonlinear tail risk functional under arbitrary data-generating processes that may drift or shift strategically over time. By leveraging deep connections between conformal tail risk control, online learning, and the variational representation of CVaR introduced by Rockafellar and Uryasev, we develop a novel procedure for online CVaR control with adversarial regret guarantees. The proposed method operates without assumptions on the underlying data-generating process, making it broadly applicable in modern high-stakes deployment settings. We prove that the realized empirical CVaR is asymptotically controlled at the target level, and that the resulting control is asymptotically tight up to a finite-sample conservatism gap. We demonstrate the effectiveness of our approach on portfolio risk management and toxicity mitigation for Large Language Models (LLMs), where rare but catastrophic failures dominate system risk.", "date": "2026-05-29", "categories": []} {"id": "2606.00690", "title": "DistMatch: Adaptive Binning via Distribution Matching for Robust Sequential Conformal Prediction", "abstract": "Sequential conformal prediction (CP) provides valid uncertainty quantification under the assumption of residual exchangeability. However, this assumption is often violated in real-world time series due to temporal dependencies and distributional shifts. While recent methods attempt to approximate exchangeability through reweighting, identifying optimal weights remains an open challenge. To address this limitation, we propose DistMatch, a binning-based method that recursively partitions residuals within a binary tree using the Kolmogorov-Smirnov (KS) statistic. We theoretically show that this partitioning induces approximately exchangeable leaves, thereby avoiding the need for reweighting. By applying quantile regression with online updates within each leaf, DistMatch enables locally adaptive inference and improves robustness to distributional shifts. Extensive experiments demonstrate that DistMatch outperforms existing sequential CP methods.", "date": "2026-05-30", "categories": []} {"id": "2606.00717", "title": "Multi-Agent Conformal Prediction with Personalized Statistical Validity", "abstract": "Uncertainty quantification is essential in high-stakes machine learning tasks. However, one of the principled solutions, conformal prediction, faces challenges under limited local calibration data, privacy constraints, and data heterogeneity. In multi-agent settings, existing works do not simultaneously and satisfactorily address these challenges with guarantees either limited to averages across agents or losing validity in heterogeneous settings. Hence, we propose personalized federated weighted conformal prediction (PFWCP), a framework that combines local density ratio weighting with weighted quantile aggregation to correct for heterogeneity while preserving privacy. The method yields asymptotically valid marginal and calibration-conditional coverage guarantees for each participating agent and supports protocols with one-shot communication. Theoretical analysis presents an adjustment to the coverage variance, governed by an effective sample size expression, which is necessary in the context of weighted conformal prediction, and experiments on synthetic and real datasets show improved calibration quality over state-of-the-art federated conformal baselines.", "date": "2026-05-30", "categories": []} {"id": "2507.17026", "title": "Conformal C2ST: Turning weak classifiers into strong two-sample tests", "abstract": "The two-sample testing problem, a fundamental task in statistics and machine learning, seeks to determine whether two sets of samples, drawn from underlying distributions $p$ and $q$, are in fact identically distributed (i.e. whether $p=q$). A popular and intuitive approach is the classifier two-sample test (C2ST), where a classifier is trained to distinguish between samples from $p$ and $q$. Yet despite simplicity of the C2ST, its reliability hinges on access to a near-Bayes-optimal classifier, a requirement that is rarely met and difficult to verify. This raises a major open question: can a weak classifier still be useful for two-sample testing? We show that the answer is a definitive yes. Building on the work of Hu and Lei (2024), we analyze two conformal variants of the C2ST that convert the scores from any trained classifier -- even if weak, biased, or overfit -- into exact, finite-sample p-values. We establish two key theoretical properties of the conformal C2ST: (i) finite-sample Type-I error control, and (ii) non-trivial power that degrades gently in tandem with the error of the trained classifier. The upshot is that even poorly performing classifiers can yield powerful and reliable two-sample tests. This general framework finds a powerful application in Bayesian inference, particularly for validating Neural Posterior Estimation (NPE) models, where the task of comparing a learned posterior approximation $q(\\theta \\mid y)$ to the true posterior $p(\\theta \\mid y)$ can be framed as a two-sample test. Empirically, the Conformal C2ST outperforms classical discriminative tests across a wide range of benchmarks for this task. Our results establish the conformal C2ST as a practical, theoretically grounded diagnostic tool.", "date": "2025-07-22", "categories": []} {"id": "2512.11779", "title": "Conditional Coverage Diagnostics for Conformal Prediction", "abstract": "Evaluating conditional coverage remains one of the most persistent challenges in assessing the reliability of predictive systems. Although conformal methods can give guarantees on marginal coverage, no method can guarantee to produce sets with correct conditional coverage, leaving practitioners without a clear way to interpret local deviations. To overcome sample-inefficiency and overfitting issues of existing metrics, we cast conditional coverage estimation as a classification problem. Conditional coverage is violated if and only if some classifier can achieve lower risk than the target coverage. Through the choice of a (proper) loss function, the resulting risk difference gives a conservative estimate of natural miscoverage measures such as L1 and L2 distance, and can even separate the effects of over- and under-coverage, and non-constant target coverages. We call the resulting family of metrics excess risk of the target coverage (ERT). We show experimentally that the use of modern classifiers provides much higher statistical power than simple classifiers underlying established metrics like CovGap. Additionally, we use our metric to benchmark different conformal prediction methods. Finally, we release an open-source package for ERT as well as previous conditional coverage metrics. Together, these contributions provide a new lens for understanding, diagnosing, and improving the conditional reliability of predictive systems.", "date": "2025-12-12", "categories": []} {"id": "2602.13110", "title": "SCOPE: Selective Conformal Optimized Pairwise LLM Judging", "abstract": "Large language models (LLMs) are increasingly used as scalable judges in pairwise evaluation, but they remain prone to miscalibration and biases. We propose SCOPE (Selective Conformal Optimized Pairwise Evaluation), a framework that calibrates an acceptance threshold so that, under exchangeability, the error rate among non-abstained judgments is at most a user-specified level $\\alpha$. To supply SCOPE with a bias-neutral uncertainty signal, we introduce Bidirectional Preference Entropy (BPE), which queries the judge under both response positions and converts the order-averaged preference probability into an entropy-based score. Across various pairwise judging benchmarks, BPE outperforms standard confidence proxies in calibration and discrimination, while SCOPE consistently satisfies the target risk bound (empirical FDR $\\approx 0.097$ to $0.099$ at $\\alpha = 0.10$) and retains substantial coverage. Compared to vanilla baselines, SCOPE accepts up to $2.4\\times$ more judgments under the same risk constraint, demonstrating that BPE enables reliable and high-coverage LLM-based evaluation.", "date": "2026-02-13", "categories": []} {"id": "2509.24100", "title": "SpeedCP: Fast Kernel-based Conditional Conformal Prediction", "abstract": "Conformal prediction provides distribution-free prediction sets with finite-sample conditional guarantees. We build upon the RKHS-based framework of Gibbs et al. (2023), which leverages families of covariate shifts to provide approximate conditional conformal prediction intervals, an approach with strong theoretical promise, but with prohibitive computational cost. To bridge this gap, we develop a stable and efficient algorithm that computes the full solution path of the regularized RKHS conformal optimization problem, at essentially the same cost as a single kernel quantile fit. Our path-tracing framework simultaneously tunes hyperparameters, providing smoothness control and data-adaptive calibration. To extend the method to high-dimensional settings, we further integrate our approach with low-rank latent embeddings that capture conditional validity in a data-driven latent space. Empirically, our method provides reliable conditional coverage across a variety of modern black-box predictors, improving the interval length of Gibbs et al. (2023) by 30%, while achieving a 40-fold speedup.", "date": "2025-09-28", "categories": []} {"id": "2605.28920", "title": "Conf-Gen: Conformal Uncertainty Quantification for Generative Models", "abstract": "Conformal prediction (CP) and its extension, conformal risk control (CRC), are established frameworks for quantifying uncertainty in supervised machine learning through formal guarantees. However, recent breakthroughs in artificial intelligence (AI) have been driven by unsupervised generative models, such as large language models (LLMs) and image generators, which are not directly compatible with CP or CRC. In this work we introduce conformal generation (Conf-Gen), a general framework adapting CRC to generative tasks while relaxing its theoretical assumptions. Conf-Gen unifies and generalizes previous attempts to apply CP to LLMs, and extends conformal methodology to entirely new domains. We demonstrate the flexibility of Conf-Gen through some novel applications, including obtaining conformal guarantees on: image generators producing non-memorized images, conversational AI systems having asked enough clarifying questions, and the output of AI agents being correct.", "date": "2026-05-27", "categories": []} {"id": "2605.29139", "title": "Anytime-Valid Federated Conformal RAG for LLM Swarms", "abstract": "Federated Conformal RAG (FC-RAG) provides distribution-free coverage for a bandwidth-limited swarm of weak language models, but only at a fixed horizon. We extend it to anytime-valid sequential coverage: validity at every stopping time, preserved under predictable adaptive control (recalibration, per-node bandwidth escalation, distilled-student refresh), at no extra cost in assumptions over fixed-horizon FC-RAG. Naive composition fails because FC-RAG's marginal coverage bound makes the betting e-process a non-supermartingale on adverse calibration draws, and Ville's inequality cannot be invoked. We give Anytime-FC-RAG, a sequential extension built on a summable per-step calibration-deviation budget that converts the marginal bound into a strict conditional bound on a calibration-good event, paired with a truncated betting e-process that is a nonnegative supermartingale on the entire probability space. From these two ingredients, we obtain four guarantees: time-uniform alarm validity $\\mathbb{P}(\\sup_t E_t \\ge 1/\\delta_e) \\le \\delta_e + \\delta_{\\mathrm{cal}}$, a Hoeffding-stitched cumulative-miscoverage envelope at the same total budget, safety under any predictable controller (recalibration, bandwidth escalation, student refresh), and training-side error propagation across an unbounded sequence of Federated Probe-Logit Distillation (FPLD) refreshes via a summable training budget. As a practical consequence, an adaptive controller that escalates retrieval bandwidth only when the e-process crosses a warning threshold matches the alarm rate of a fixed-high-bandwidth schedule at substantially lower communication cost. Experiments on a GPT-2-small + MiniLM swarm across MMLU, DBpedia, and AG News verify the predicted alarm rate, detection delay, envelope coverage, and $14$-$57\\%$ bandwidth savings; the alarm fires when and only when coverage genuinely breaks.", "date": "2026-05-27", "categories": []} {"id": "2605.30085", "title": "Conformal Certification of Reasoning Trace Prefixes", "abstract": "Language model reasoning traces are rarely all-or-nothing; they frequently contain valid intermediate steps before a critical error occurs. Existing uncertainty quantification methods typically certify final answers or entire responses, failing to provide statistical guarantees for the proportion of a sequential trace that can be safely retained. To address this, we introduce CROP (Conformal Reasoning Output Prefixes), a verifier-agnostic calibration procedure for clean-prefix certification. Given any step-level risk proxy, CROP selects a calibrated threshold and returns the longest contiguous prefix whose step risk proxies remain below it, routing the uncertified suffix for downstream review or repair. Assuming exchangeability, CROP rigorously controls the marginal probability that the returned prefix contains an annotated error. Across six process-labeled reasoning datasets, we demonstrate that standard step-level metrics such as AUROC do not fully capture prefix utility, suggesting verifiers should instead be evaluated by certified prefix length. Furthermore, CROP balances over- and under-withholding, improving downstream repair accuracy by preserving valid intermediate reasoning while discarding misleading suffixes. Ultimately, this work positions prefix certification as a rigorous, practical bridge between process supervision, abstention, and repair.", "date": "2026-05-28", "categories": []} {"id": "2411.13479", "title": "Conformal Prediction for Hierarchical Data", "abstract": "We consider conformal prediction for multivariate data and focus on hierarchical data, where some components are linear combinations of others. Intuitively, the hierarchical structure can be leveraged to reduce the size of prediction regions for the same coverage level. We implement this intuition by including a projection step (also called a reconciliation step) in the split conformal prediction [SCP] procedure, and prove that the resulting prediction regions are indeed globally smaller. We do so both under the classic objective of joint coverage and under a new and challenging task: component-wise coverage, for which efficiency results are more difficult to obtain. The associated strategies and their analyses are based both on the literature of SCP and of forecast reconciliation, which we connect. We also illustrate the theoretical findings, for different scales of hierarchies on simulated data.", "date": "2024-11-20", "categories": []} {"id": "2605.26429", "title": "Structure-Adaptive Conformal Inference for Large-Scale Out-of-Distribution Testing", "abstract": "This paper addresses structured out-of-distribution (OOD) testing in high-stakes machine learning applications. Traditional conformal methods rely on joint exchangeability, making it difficult to incorporate auxiliary information such as spatiotemporal or grouping structures. To overcome this limitation, we propose the structure-adaptive conformal q-value (SCQ), a significance index that integrates individual test evidence with structural patterns. We also develop pseudo-score-guided transductive automated model selection (P-TAMS), which adapts conformalized model selection to structured OOD testing across a toolbox of candidate models. Together, SCQ and P-TAMS form a unified framework under pairwise exchangeability, providing finite-sample error-rate control, improved power, and enhanced interpretability. Experiments on simulated and real data demonstrate that the proposed approach controls the false discovery rate and performs well across diverse settings.", "date": "2026-05-26", "categories": []} {"id": "2605.26569", "title": "Distribution-Aware Conformal Prediction: A Framework for generating efficient prediction intervals for time series", "abstract": "We present Distribution-aware Conformal Prediction (DCP), a unified framework integrating probabilistic predictors like Monte Carlo dropout, deep ensembles, and quantile regression with score-agnostic conformal calibration to produce valid and efficient prediction intervals. Leveraging a numerical inversion approach to construct interval bounds, DCP accommodates arbitrary combinations of distribution generating predictors and nonconformity scores. Benchmark analysis on synthetic and real-world time series data demonstrate DCP's ability to adaptively calibrate prediction intervals under varying uncertainty regimes. Crucially, DCP's modular design facilitates plug-and-play experimentation with different predictor-score pairings, quantitatively supported by a newly introduced modified Winkler score that balances validity and efficiency by explicitly penalizing undercoverage. While DCP generalizes and extends existing approaches like Conformalized Quantile Regression and Conformalized Monte Carlo, its modular design allows further extensions, setting a foundation for advancing uncertainty quantification in dynamic environments and high-risk applications.", "date": "2026-05-26", "categories": []} {"id": "2605.27091", "title": "MiRD: Reliable Set-Valued Prediction for Open-Ended Question Answering via Miscoverage Risk Decomposition", "abstract": "Reliable set-valued prediction provides a principled way to mitigate hallucinations in open-ended question answering (QA), yet existing conformal approaches typically rely on a fragile premise: finite sampling must already produce at least one admissible candidate, or calibration examples violating this condition are discarded. In this paper, we introduce MiRD, a two-stage framework that decomposes overall miscoverage into sampling failure and conditional selection failure. In Stage I, MiRD establishes an expectation-level marginal upper bound on the probability that finite sampling produces no admissible answer under a fixed budget. In Stage II, conditioned on sampling success, MiRD calibrates a conformal selection threshold using admission-correlated nonconformity scores defined over the full calibration set, thereby preserving calibration-set integrity. Across three open-ended QA datasets and eight models, MiRD controls sampling risk, conditional selection risk, and overall miscoverage, while yielding tighter first-stage bounds than PAC-style alternatives and more adaptive prediction sets than successful-only calibration.", "date": "2026-05-25", "categories": []} {"id": "2605.24872", "title": "Cluster Frequency Conformal Prediction for Local Coverage", "abstract": "Conformal prediction provides distribution-free coverage guarantees, but in many-class classification it may still under-cover specific classes or subpopulations, preventing safe deployment in high-stakes applications. We propose Cluster Frequency Conformal Prediction (CFCP), a plug-in framework that adapts conformal prediction to local structure in a learned representation space. CFCP clusters learned embeddings, estimates cluster-level label-frequency distributions from calibration data, and for each test point constructs a sample-specific probability vector by softly mixing nearby cluster distributions regularized with global-prior and reliability-aware shrinkage. This vector is then conformalized using standard set constructors. In the disjoint-split regime, CFCP inherits standard finite-sample marginal validity. Under additional assumptions, CFCP further admits a local-validity interpretation. Since representation clusters aggregate locally similar samples, their empirical class frequencies provide a stable estimate of local label ambiguity. Across image and text benchmarks, CFCP achieves the best class coverage in 15/16 dataset/score-family comparisons and a competitive prediction set size efficiency, with several settings substantially more efficient. Overall, our results show that cluster-frequency information provides an effective localized signal for improving classwise reliability in many-class conformal prediction.", "date": "2026-05-24", "categories": []} {"id": "2605.24983", "title": "Benchmarking non-conformity score functions in conformal prediction", "abstract": "Conformal prediction is a useful and versatile alternative to model calibration in machine learning classification. It replaces single-class prediction with prediction sets, guaranteeing that the \\textit{a priori} probability of the prediction sets containing the true class is larger than or equal to a pre-specified rate. The size and usefulness of the prediction sets relies heavily on the choice of the non-conformity score function. The scientific literature contains many examples of non-conformity score functions but there is an absence of studies examining their properties and effectiveness. In this paper, we give an overview of properties of non-conformity score functions. We give examples of non-conformity score functions in the existing literature and introduce original modifications. We introduce an original method of evaluating the prediction set sizes of conformal predictors and use it to provide a comparison between non-conformity score functions. We also examine efficacy of different non-conformity score functions for class-conditional conformal prediction in a setting with imbalanced classes.", "date": "2026-05-24", "categories": []} {"id": "2605.25882", "title": "Conformalised imprecise inference for robust extrapolation under limited data", "abstract": "Recent advances in uncertainty quantification increasingly emphasise the distinction between aleatory and epistemic uncertainty in machine learning, motivating the need for more unified frameworks. However, despite much progress in producing reliable predictions, existing methods often lack rigorous guarantees when generalising beyond the training domain. We propose a conformalised imprecise inference framework for robust extrapolation, which is model-agnostic and augments predictive models with imprecision and distance awareness. The proposed approach yields imprecise predictions (probability boxes) that remain valid under distributional shift, maintaining coverage while adaptively expanding uncertainty in extrapolation regimes. Experiments on synthetic and benchmark datasets demonstrate improved robustness and reliable coverage compared to standard probabilistic approaches, particularly under limited data.", "date": "2026-05-25", "categories": []} {"id": "2510.04406", "title": "Decomposition-Based Modular Conformal Prediction for Two-Stage Modeling", "abstract": "Conformal prediction offers finite-sample coverage guarantees under minimal assumptions. However, existing methods treat the entire modeling process as a black box, overlooking opportunities to exploit and understand modular structure. We introduce a conformal prediction framework for two-stage sequential models, where an upstream predictor generates intermediate representations for a downstream model. By decomposing the overall prediction residual into stage-specific components, our method enables practitioners to attribute uncertainty to specific pipeline stages. We develop a risk-controlled parameter selection procedure using family-wise error rate (FWER) control to calibrate stage-wise scaling parameters, and introduce an adaptive extension for non-stationary settings. Experiments on synthetic distribution shifts, as well as real-world supply chain and stock market data, demonstrate that our approach improves coverage under structural, stage-wise shifts compared to standard conformal methods, while identifying stage-wise error contribution. This framework offers diagnostic advantages and robust coverage that standard conformal methods lack.", "date": "2025-10-06", "categories": []} {"id": "2605.23189", "title": "Empirical Bayes Conformal Prediction for Vision and Language Models", "abstract": "Conformal prediction (CP) gives distribution-free coverage for modern vision and language models, but it is often forced to make a ranking decision from a single unstable nonconformity score. Standard CP uses one realization, while average-then-calibrate variants smooth multiple realizations into a point estimate. Both options discard the inconsistency that can help identify whether a candidate is indeed stable. A weak answer can enter the conformal set even if the evidence is not strong, simply because one posterior sample or prompt phrasing made it look strong. But variability can help distinguish a stable signal from noise-driven fluctuations. We describe an empirical Bayes conformal prediction framework that uses $r$-values to convert score variability into an uncertainty informed nonconformity score. The resulting $r$-value estimates how likely a candidate's latent score belongs to the top-ranked group after accounting for both its mean score and its uncertainty. It admits both a closed-form Normal-Normal empirical Bayes estimator and a nonparametric posterior-sampling estimator. Using the $r$-value as the nonconformity score preserves the target conformal coverage while provably reducing the inclusion of high variance false candidates under mild regularity conditions. Across image classification, CLIP-based VLM benchmarks, and LLMs, we show that $r$-value conformal prediction preserves target coverage while improving ranking stability and reducing set size when variability is informative, and reverting to CP-like behavior when variability vanishes.", "date": "2026-05-22", "categories": []} {"id": "2605.21543", "title": "Provable Joint Decontamination for Benchmarking Multiple Large Language Models", "abstract": "Benchmark data contamination has become a central challenge in LLM evaluation: when evaluation examples appear in the training data of one or more audited models, reported performance can be inflated and cross-model comparisons become unreliable. A broad line of training-data detection work designs scores to quantify how strongly a model memorizes a given data point, but these score-based methods lack theoretical guarantees. Recent conformal approaches provide provable false-identification control for a single model; however, applying them separately to each model can produce model-specific benchmarks, undermining fair comparison across models. In this work, we formalize multi-model benchmark decontamination as a joint selection problem and propose Joint Envelope Conformal Selection (JECS), a conformal procedure that enables global contamination rate (GCR) control under stated assumptions. Specifically, JECS computes per-model conformal p-values, aggregates them by the per-item maximum, and reconstructs a conservative envelope of the max-p null distribution from right-tail observations above a data-driven threshold. By applying the adaptive Benjamini-Hochberg (BH) procedure to the envelope-rescaled values, we select a benchmark with provable GCR control. Extensive experiments across various models and benchmarks demonstrate that JECS achieves higher power than the max-p baseline while consistently maintaining the target GCR control.", "date": "2026-05-20", "categories": []} {"id": "2605.20270", "title": "Conformal Selective Acting: Anytime-Valid Risk Control for RLVR-Trained LLMs", "abstract": "A local specialist LLM, fine-tuned with reinforcement learning from verifiable rewards (RLVR) on operator-local data, is installed in a regulated organization with per-deployment error budget $\\alpha$. The operator needs a safety certificate for this deployment's stream at every round: no pooling across deployments, no waiting for a long-run average. Existing wrappers cannot deliver this on adaptive, online-updated streams: offline conformal-risk methods require exchangeability; online-conformal methods bound only long-run averages; non-exchangeable extensions are marginally valid; and the closest anytime wrapper, A-RCPS, controls marginal rather than selective risk. Using a (test statistic, validity guarantee, deployment rule) framework, we identify one empty cell forced by deployment requirements: e-process per threshold, selective risk, anytime-pathwise validity, max-certified-threshold rule. Conformal Selective Acting (CSA) fills it as a per-round wrapper maintaining a Ville-type e-process per threshold on a Bonferroni grid, evaluated against the RLVR filtration. Under predictable updates and isotonic-calibrated monotone risk we prove (i) an anytime-pathwise selective-risk bound $R_T^{\\mathrm{act}}\\le\\alpha+O(N_T^{-1/2})$, (ii) rate-optimal certification matching $\\Theta(\\bar\\eta^{-2}\\log(1/\\delta))$, and (iii) a horizon-independent release-rate gap. Across eight specialist benchmarks ($480$ streams), sixteen adversarial distribution-shift cells ($160$ streams), and five live Expert-Iteration RLVR cells with online LoRA over four base models in three architecture families ($10{,}300$ rounds), CSA is the only method among ten compared that satisfies pathwise validity and non-refusing deployment on every cell. We do not propose a new LLM, training algorithm, or policy class; CSA is the deployment-side complement, orthogonal to the model, for operators who cannot use a frontier API.", "date": "2026-05-18", "categories": []} {"id": "2605.20515", "title": "Online Conformal Prediction with Corrupted Feedback", "abstract": "Modern artificial intelligence systems require calibrated uncertainty estimates that remain reliable in sequential and non-stationary environments. Online conformal prediction (OCP) addresses this challenge through adaptively updated prediction sets that provide deterministic long-run miscoverage guarantees. These guarantees, however, hinge on the assumption of perfect feedback about the coverage of past prediction sets. In practice, the observed miscoverage indicator may be corrupted by noise, communication failures, or adversarial manipulation, which can severely degrade OCP's calibration guarantees. In this paper, we study OCP under corrupted feedback. We first model feedback corruption as an arbitrary binary flip sequence, and analyze how feedback corruption affects and degrades the miscoverage performance of standard OCP. We then propose two robust schemes: robust OCP via filtering, which leverages the structural properties of the predicted threshold to filter corrupted feedback, and robust OCP via active compensation, which incorporates an active compensation mechanism to mitigate the effect of corrupted feedback. For both methods, we establish explicit miscoverage guarantees, which are further specialized for an independent stochastic flip model and for an arbitrary error model with memory bounds. Experiments on real-world datasets validate the proposed approach, showing markedly improved calibration and significantly smaller prediction sets compared with baseline OCP methods under corrupted feedback.", "date": "2026-05-19", "categories": []} {"id": "2605.20956", "title": "A Deployment Audit of Release-Side Risk in Conformal Triage under Prevalence Shift", "abstract": "Conformal triage converts predictive scores into deployment actions that either release a case, flag it for urgent attention, or defer it to human review. Under prevalence shift, however, the usual summaries of marginal coverage and human-review rate can miss the safety-critical question of whether patients who truly experience the target event are released without review. To address this gap, we introduce a leakage-aware deployment audit for release-side conformal triage. It first assigns target subjects to three non-overlapping roles: prevalence correction, conformal calibration, and held-out release-safety evaluation. This separation then lets the audit evaluate release directly: how many event-positive patients are cleared without review, whether the pilot has enough event labels for calibration, and how the safety-review trade-off shifts. Applying this audit to a retrospective NSCLC pilot shows why lower review can be misleading: after prevalence correction, the pooled conformal branch lowers review by releasing more patients, some of whom are event-positive. Within the audit, the classwise branch acts as a scarcity diagnostic: the pilot has too few event labels to certify safe low-review release.", "date": "2026-05-20", "categories": []} {"id": "2512.24139", "title": "Colorful Pinball: Density-Weighted Quantile Regression for Conditional Guarantee of Conformal Prediction", "abstract": "Although conformal prediction provides robust marginal coverage guarantees, achieving reliable conditional coverage for specific inputs remains challenging. While exact distribution-free conditional coverage is impossible with finite samples, recent work has focused on improving the conditional coverage of standard conformal procedures. Distinct from approaches that target relaxed notions of conditional coverage, we directly target the mean squared error of conditional coverage by refining the quantile regression components that underpin many conformal methods. Leveraging a Taylor expansion, we derive a sharp surrogate objective for quantile regression: a density-weighted pinball loss, where the weights are given by the conditional density of the nonconformity score evaluated at the true quantile. We propose a three-headed quantile network that estimates these weights via finite differences using auxiliary quantile levels at $1-\\alpha \\pm \\delta$, subsequently fine-tuning the central quantile by optimizing the weighted loss. We provide a theoretical analysis with exact non-asymptotic guarantees characterizing the resulting excess risk. Extensive experiments on diverse high-dimensional real-world datasets demonstrate remarkable improvements in conditional coverage performance.", "date": "2025-12-30", "categories": []} {"id": "2605.18812", "title": "PASC: Pipeline-Aware Conformal Prediction with Joint Coverage Guarantees for Multi-Stage NLP and LLM Pipelines", "abstract": "Modern NLP and LLM systems are pipelines: named entity recognition (NER) -> entity disambiguation (NED) -> entity typing, retrieval-augmented generation (retriever -> reader), and agentic chains of planner -> tool -> critic. Errors compound across stages, but existing uncertainty quantification methods either calibrate each stage independently (no joint coverage) or apply a Bonferroni union bound (joint coverage, but conservative). We present PASC (Pipeline-Aware Split Conformal), which reduces multi-stage joint coverage to a single scalar conformal prediction problem on the joint maximum nonconformity score. PASC provides a finite-sample distribution-free guarantee that all K stages are simultaneously covered with probability at least 1 - alpha, and is nearly tight up to a 1/(n+1) factor. On a three-stage NER -> NED -> entity-typing pipeline over CoNLL-2003, PASC achieves 96.4% end-to-end coverage versus 93.4% for Bonferroni and 86.5% for independent CP, at identical average prediction set size (1.083). Under distribution shift to WNUT-17 Twitter and WikiNEuRal Wikipedia data, PASC empirically maintains the target coverage in the tested shift settings while independent CP collapses to 59%. PASC requires a single quantile computation, runs 1.7x faster than Bonferroni, and scales to K = 6 stages where independent CP drops to 0.53 end-to-end coverage. The same joint-maximum-score reduction applies directly to compound LLM systems and agent pipelines.", "date": "2026-05-12", "categories": []} {"id": "2605.19024", "title": "Conformal Prediction via Transported Beta Laws", "abstract": "Split conformal prediction provides finite-sample marginal coverage under exchangeability, but this guarantee averages over the random calibration sample. We study instead the law of the calibration-conditional coverage induced by a realized conformal threshold. In the continuous i.i.d. setting this law is exactly $Beta(k,n+1-k)$, so the usual marginal guarantee corresponds to its mean. We take this beta law as a finite-sample reference object and quantify departures from it using Wasserstein distances on $[0,1]$. The framework yields direct bounds on marginal coverage gaps and on bad-calibration probabilities, and separates different sources of non-i.i.d. behavior according to how they deform the beta reference: test-side shift acts through a transport map on the coverage scale, while calibration dependence changes the order-statistic law itself. We instantiate the framework in scale-shift, clustered, and stationary mixing settings, where the induced deformations can be characterized explicitly or through Berry-Esseen approximations. Simulations on dependent processes confirm that the first-order approximation tracks the empirical Wasserstein distance even at moderate sample sizes.", "date": "2026-05-18", "categories": []} {"id": "2605.19779", "title": "Distribution-Free Uncertainty Quantification for Continuous AI Agent Evaluation", "abstract": "We adapt split conformal prediction and adaptive conformal inference (ACI) to continuous AI agent evaluation, providing distribution-free coverage guarantees for forecasted quality scores. Conformal intervals achieve calibration error below 0.02 across all nominal levels at the 24h horizon, while ACI correctly widens intervals by 35% following agent releases then reconverges. We further develop compositional uncertainty bounds for multi-agent pipelines (validated via simulation across inter-stage correlations rho in [-0.5, 0.9]), a conformal abstention rule for pairwise rankings with controlled false-ranking rate, and FDR-corrected abstention for leaderboard-scale multiple testing. Evaluating 50 agents via 18 real-time signals collected hourly, we show that per-agent conditional coverage is well-concentrated around the nominal level (mean 80.4%, 90% of agents within [72%, 90%]), and that cross-source sentiment divergence predicts ranking instability (r=0.64, p<0.01). A circularity-controlled validation confirms the framework captures signal beyond benchmarks (rho_s=0.52, p<0.01, n=35). Code and data are released under CC BY 4.0.", "date": "2026-05-19", "categories": []} {"id": "2602.01733", "title": "ST-BCP: Tightening Coverage Bound for Backward Conformal Prediction via Non-Conformity Score Transformation", "abstract": "Conformal Prediction (CP) provides a statistical framework for uncertainty quantification that constructs prediction sets with coverage guarantees. While CP yields uncontrolled prediction set sizes, Backward Conformal Prediction (BCP) inverts this paradigm by enforcing a predefined upper bound on set size and estimating the resulting coverage guarantee. However, the looseness induced by Markov's inequality within the BCP framework causes a significant gap between the estimated coverage bound and the empirical coverage. In this work, we introduce ST-BCP, a novel method that introduces a data-dependent transformation of nonconformity scores to narrow the coverage gap. In particular, we develop a computable transformation and prove that it outperforms the baseline identity transformation. Extensive experiments demonstrate the effectiveness of our method, reducing the average coverage gap from 4.20\\% to 1.12\\% on common benchmarks.", "date": "2026-02-02", "categories": []} {"id": "2605.17705", "title": "Online Conformal Prediction for Non-Exchangeable Panel Data", "abstract": "Panel data, in which multiple units are repeatedly observed over time, arise throughout science and engineering. Quantifying predictive uncertainty in such settings is challenging because conformal prediction, while distribution-free and model-agnostic, classically relies on exchangeability assumptions that fail under temporal dependence and unit heterogeneity. We propose a simple online conformal framework for non-exchangeable panel data. The method exploits a key feature of online panel prediction: when a forecast is required for one unit, contemporaneous outcomes from related units may already be observed and can serve as a calibration panel. At each round, prediction sets are formed using currently observed calibration units together with two adaptive quantities: history-based similarity weights that emphasize calibration units resembling the target, and an adaptive miscoverage level that is updated whenever target feedback is revealed. This two-state design yields a stepwise coverage bound and a long-run coverage guarantee. Empirically, across synthetic and real panel data sets, the method improves coverage on the worst-covered target units through adaptive interval-width allocation rather than uniform inflation. The two states are complementary: similarity weights protect coverage when target feedback is sparse, while the adaptive level further improves coverage as feedback accumulates.", "date": "2026-05-18", "categories": []} {"id": "2605.18202", "title": "Concise and Logically Consistent Conformal Sets for Neuro-Symbolic Concept-Based Models", "abstract": "Neuro-Symbolic Concept-based Models (NeSy-CBMs) are a family of architectures that integrate neural networks with symbolic reasoning for enhanced reliability in high-stakes applications. They work by first extracting high-level concepts from the input and then inferring a task label from these compatibly with given logical constraints. Yet, their label and concept predictions can be overconfident, making it difficult for stakeholders to gauge when the model's decisions can be trusted. We address this issue by integrating ideas from Conformal Prediction (CP), a framework providing rigorous, distribution-free coverage guarantees. We formalize three desiderata -- consistency, coverage, and conciseness -- that any conformal method for NeSy-CBMs should satisfy, and show that existing approaches fall short of at least one. We then introduce COCOCO, a post-hoc framework that conformalizes concepts and labels jointly and reconciles them via a single deduction-abduction revision step. COCOCO satisfies all three desiderata, retains distribution-free coverage, is robust to imperfect knowledge and supports user-specified size budgets. Our experiments on 8 data sets highlight how COCOCO compares favorably against competitors and natural baselines in terms of performance and set size.", "date": "2026-05-18", "categories": []} {"id": "2605.18354", "title": "Decoupled Conformal Optimisation: Efficient Prediction Sets via Independent Tuning and Calibration", "abstract": "Bayesian conformal optimisation methods often use the same held-out data both to search for efficient prediction sets and to certify coverage or risk. This coupling is natural for high-probability risk-control guarantees, but it is not necessary when the target is standard finite-sample marginal conformal coverage. We propose Decoupled Conformal Optimisation (DCO), a train-tune-calibrate design principle that uses an independent tuning split for efficiency-oriented structural selection and a fresh calibration split for the final conformal quantile. Conditional on the tuned structure, standard split-conformal exchangeability yields finite-sample marginal coverage for any candidate class, without a confidence parameter or multiple-testing correction. DCO therefore targets a different finite-sample guarantee from PAC-style methods: marginal conformal coverage rather than high-probability risk control. Under consistency assumptions on the coupled risk bound, the two approaches nevertheless converge to the same population threshold. Across classification and regression benchmarks, including ImageNet-A, CIFAR-100, Diabetes, California Housing, and Concrete, DCO tracks the nominal coverage level closely while often reducing average prediction-set size or interval width relative to PAC-style calibration. On ImageNet-A, for example, the average set size decreases from $26.52$ to $25.26$ and the 95th-percentile set size from $58.95$ to $53.73$; on Diabetes, the average interval width decreases from $2.098$ to $1.914$.", "date": "2026-05-18", "categories": []} {"id": "2511.18225", "title": "Adaptive Conformal Prediction for Quantum Machine Learning", "abstract": "Quantum machine learning seeks to leverage quantum computers to improve upon classical machine learning algorithms. Currently, robust uncertainty quantification methods remain underdeveloped in the quantum domain, despite the critical need for reliable and trustworthy predictions. Recent work has introduced quantum conformal prediction, a framework that produces prediction sets that are guaranteed to contain the true outcome with a user-specified probability. In this work, we formalise how the time-varying noise inherent in quantum processors can undermine conformal guarantees, even when calibration and test data are exchangeable. To address this challenge, we draw on Adaptive Conformal Inference, a method which maintains validity over time via repeated recalibration. We introduce Adaptive Quantum Conformal Prediction (AQCP), an algorithm which provides asymptotic average coverage guarantees under arbitrary hardware noise conditions. Empirical studies on an IBM quantum processor demonstrate that AQCP achieves the target coverage level and exhibits greater stability than quantum conformal prediction.", "date": "2025-11-23", "categories": []} {"id": "2604.15221", "title": "Vision-Based Safe Human-Robot Collaboration with Uncertainty Guarantees", "abstract": "We propose a framework for vision-based human pose estimation and motion prediction that gives conformal prediction guarantees for certifiably safe human-robot collaboration. Our framework combines aleatoric uncertainty estimation with OOD detection for high probabilistic confidence. To integrate our pipeline in certifiable safety frameworks, we propose conformal prediction sets for human motion predictions with high, valid confidence. We evaluate our pipeline on recorded human motion data and a real-world human-robot collaboration setting.", "date": "2026-04-16", "categories": []} {"id": "2605.14260", "title": "On the Burden of Achieving Fairness in Conformal Prediction", "abstract": "Conformal prediction is often calibrated with a single pooled threshold, but this can hide cross-group heterogeneity in score distributions and distort group-wise coverage. We study this phenomenon through the population score distributions underlying split conformal calibration. First, we derive a conservation law and lower bound showing that pooled calibration incurs irreducible group-wise coverage distortion at a scale set by cross-group quantile heterogeneity. Second, we demonstrate that the two leading fairness definitions for conformal prediction, Equalized Coverage and Equalized Set Size, are fundamentally in tension. Third, we quantify the cost of moving between policies which treat groups separately or pool them. Experiments on synthetic and real data confirm the same bidirectional trade-off after finite-sample calibration. Our results show that, for the policy families studied here, calibration choice does not remove cross-group heterogeneity; it determines whether the resulting distortion appears in the coverage or size dimension, providing a principled lens for analyzing fairness-oriented calibration choices in practice.", "date": "2026-05-14", "categories": []} {"id": "2605.16145", "title": "Skew-adaptive conformal prediction", "abstract": "We develop a skew-adaptive extension of split conformal prediction for regression. The method starts from an asymmetric interval family centered at a point prediction and uses the gauge approach to deduce the conformity score induced by this family. The inverse hyperbolic sine transform of signed scaled residuals provides the training target for an additional predictive model, whose role is to learn how predictive uncertainty should tilt across the feature space. The resulting procedure preserves the finite-sample marginal validity of split conformal prediction under exchangeability, while producing intervals that adapt to both local scale and local skewness. We also develop a calibration-sample-based estimator for comparing the expected relative future width of the skew-adaptive and classical scaled-score intervals. Experiments on a variety of datasets indicate gains in prediction interval efficiency over the scaled-score construction and conformalized quantile regression, and show that the proposed estimator closely matches the corresponding average width ratio observed on the test sample.", "date": "2026-05-15", "categories": []} {"id": "2410.19653", "title": "Conformal Prediction for Multimodal Regression", "abstract": "This paper introduces multimodal conformal regression. Traditionally confined to scenarios with solely numerical input features, conformal prediction is now extended to multimodal contexts through our methodology, which harnesses internal features from complex neural network architectures processing images and unstructured text. Our findings highlight the potential for internal neural network features, extracted from convergence points where multimodal information is combined, to be used by conformal prediction to construct prediction intervals (PIs). This capability paves new paths for deploying conformal prediction in domains abundant with multimodal data, enabling a broader range of problems to benefit from guaranteed distribution-free uncertainty quantification.", "date": "2024-10-25", "categories": []} {"id": "2605.14953", "title": "Efficient Online Conformal Selection with Limited Feedback", "abstract": "We address the problem of conformal selection, where an agent must select a minimal subset of options to ensure that at least one ``success'' is identified with a pre-specified target probability $\\phi$. While traditional online conformal prediction focuses on maintaining validity for the observed sequence, minimizing the resource cost (efficiency) of such selections, especially under limited feedback, remains a significant challenge. In this work, we consider settings with the most limited ``bandit'' feedback, and demonstrate that the simple Adaptive Conformal Inference (ACI) update rule, when applied to the appropriate control parameter or dual variable, is both adversarially valid, ensuring the success target is met on average for any input sequence (and hence under distribution shifts), and stochastically efficient, achieving sublinear efficiency regret for $i.i.d.$ inputs against an appropriate stochastic benchmark. We show such guarantees under canonical models capturing bandit and semi-bandit feedback to the agent via a unifying algorithmic technique, and analytic framework involving Lyapunov functions. Our approach handles more complex settings than prior work, while requiring significantly less feedback, and our results provide a new theoretical bridge between efficient online learning with limited feedback and distribution-free uncertainty quantification.", "date": "2026-05-14", "categories": []} {"id": "2602.01629", "title": "AdaptNC: Adaptive Nonconformity Scores for Conformal Prediction under Distribution Shift", "abstract": "Rigorous uncertainty quantification is essential for the safe deployment of autonomous systems in unconstrained environments. Conformal Prediction (CP) provides a distribution-free framework for this task, yet its standard formulations rely on exchangeability assumptions that are violated by the distribution shifts inherent in real-world robotics. Existing online CP methods maintain target coverage by adaptively scaling the conformal threshold, but typically employ a static nonconformity score function. We show that this fixed geometry leads to highly conservative, volume-inefficient prediction regions when environments undergo structural shifts. To address this, we propose $\\textbf{AdaptNC}$, a framework for the joint online adaptation of both the nonconformity score parameters and the conformal threshold. AdaptNC leverages an adaptive reweighting scheme to optimize score functions, and introduces a replay buffer mechanism to mitigate the coverage instability that occurs during score transitions. We evaluate AdaptNC on diverse robotic benchmarks involving multi-agent policy changes, environmental changes and sensor degradation. Our results demonstrate that AdaptNC significantly reduces prediction region volume compared to state-of-the-art threshold-only baselines while maintaining target coverage levels.", "date": "2026-02-02", "categories": []} {"id": "2605.12668", "title": "Online Conformal Prediction: Enforcing monotonicity via Online Optimization", "abstract": "Conformal prediction provides a principled framework for uncertainty quantification with finite-sample coverage guarantees. While recent work has extended conformal prediction to online and sequential settings, existing methods typically focus on a single coverage level and do not ensure consistency across multiple confidence levels. In many real-world applications, such as weather forecasting, macroeconomic prediction, and risk management, different users operate under heterogeneous risk tolerances and require calibrated uncertainty estimates across a range of coverage levels. In such settings, it is desirable to produce prediction sets corresponding to different coverage levels that are nested and valid simultaneously. In this paper, we propose two novel online conformal prediction methods that output \\emph{nested prediction sets} across a range of coverage levels, enabling simultaneous uncertainty quantification across the entire risk spectrum. Beyond interpretability, jointly estimating multiple coverage levels is known to improve statistical efficiency in classical quantile regression by enforcing non-crossing constraints and sharing information across quantiles. Our approaches leverage an online optimization perspective with small regret that translates to quantile estimation error control while enforcing nestedness of prediction sets. Empirical results on synthetic and real-world datasets, including applications in forecasting tasks with heterogeneous risk requirements, demonstrate that our method achieves stable coverage across all levels, strictly nested prediction sets, and improved efficiency compared to existing online conformal baselines.", "date": "2026-05-12", "categories": []} {"id": "2605.12917", "title": "Adaptive Conformal Prediction for Reliable and Explainable Medical Image Classification", "abstract": "Deep learning models for medical imaging often exhibit overconfidence, creating safety risks in ambiguous diagnostic scenarios. While Conformal Prediction (CP) provides distribution-free statistical guarantees, standard methods such as Regularized Adaptive Prediction Sets (RAPS) optimize for average efficiency and can mask severe failures on difficult inputs.\n We propose an Adaptive Lambda Criterion for RAPS that minimizes the worst-case coverage violation across prediction set size strata. On OrganAMNIST (58,850 abdominal CT images, 11 classes), standard size-optimized RAPS converges to near-deterministic behavior with stratified undercoverage on uncertain samples, while our method achieves 95.72 percent global coverage with average set size 1.09 and at least 90 percent coverage across all strata. Cross-domain validation on PathMNIST (107,180 pathology images, 9 classes) confirms generalizability.\n Quantitative Grad-CAM analysis (rho = -0.30, p < 1e-22) shows that multi-label predictions correspond to focused attention on anatomically ambiguous regions. These results demonstrate that the proposed method improves reliability while maintaining efficiency, making it suitable for safety-critical medical AI applications.", "date": "2026-05-13", "categories": []} {"id": "2605.13642", "title": "Conformal Anomaly Detection in Python: Moving Beyond Heuristic Thresholds with 'nonconform'", "abstract": "Most anomaly detection systems output scores rather than calibrated decisions, leaving practitioners to choose thresholds heuristically and without clear statistical interpretation. Conformal anomaly detection addresses this limitation by converting anomaly scores into calibrated p-values that are valid under the statistical assumption of data exchangeability, with a growing literature extending this idea beyond that setting. We present 'nonconform', a Python package for applying conformal anomaly detection within existing machine-learning workflows, and use it as the basis for an implementation-grounded introduction to the field. The package integrates with 'scikit-learn', 'pyod', and custom anomaly detectors, and provides a unified interface for calibration, p-value generation, and false discovery rate control. It supports several conformalization strategies, ranging from simple split-conformal calibration to more data-efficient and shift-aware extensions. Through a progression from foundational concepts to advanced conformalization strategies, complemented by code examples, the paper connects the statistical ideas behind conformal anomaly detection to their practical use in 'nonconform'. Empirical results demonstrate that the implemented methods enable statistically principled anomaly detection. Together, the package and exposition aim to make core conformal anomaly detection workflows more accessible and reproducible in experimental and production-oriented settings.", "date": "2026-05-13", "categories": []} {"id": "2603.22000", "title": "CRPS-Optimal Binning for Univariate Conformal Regression", "abstract": "We propose a method for non-parametric conditional distribution estimation based on partitioning covariate-sorted observations into contiguous bins and using the within-bin empirical CDF as the predictive distribution. Bin boundaries are chosen to minimise the total leave-one-out Continuous Ranked Probability Score (LOO-CRPS), which admits a closed-form cost function with $O(n^2 \\log n)$ precomputation and $O(n^2)$ storage; the globally optimal $K$-partition is recovered by a dynamic programme in $O(n^2 K)$ time. Minimisation of within-sample LOO-CRPS turns out to be inappropriate for selecting $K$ as it results in in-sample optimism. We instead select $K$ by $K$-fold cross-validation of test CRPS, which yields a U-shaped criterion with a well-defined minimum. Having selected $K^*$ and fitted the full-data partition, we form two complementary predictive objects: the Venn prediction band and a conformal prediction set based on CRPS as the nonconformity score, which carries a finite-sample marginal coverage guarantee at any prescribed level $\\varepsilon$. The conformal prediction is transductive and data-efficient, as all observations are used for both partitioning and p-value calculation, with no need to reserve a hold-out set. On real benchmarks against split-conformal competitors (Gaussian split conformal, CQR, CQR-QRF, and conformalized isotonic distributional regression), the method produces substantially narrower prediction intervals while maintaining near-nominal coverage.", "date": "2026-03-23", "categories": []} {"id": "2605.11684", "title": "Partial Model Sharing Improves Byzantine Resilience in Federated Conformal Prediction", "abstract": "We propose a Byzantine-resilient federated conformal prediction (FCP) method that leverages partial model sharing, where only a subset of model parameters is exchanged each round. Unlike existing robust FCP approaches that primarily harden the calibration stage, our method protects both the federated training and conformal calibration phases. During training, partial sharing inherently restricts the attack surface and attenuates poisoned updates while reducing communication. During calibration, clients compress their non-conformity scores into histogram-based characterization vectors, enabling the server to detect Byzantine clients via distance-based maliciousness scores and to estimate the conformal quantile using only benign contributors. Experiments across diverse Byzantine attack scenarios show that the proposed method achieves closer-to-nominal coverage with substantially tighter prediction intervals than standard FCP, establishing a robust and communication-efficient approach to federated uncertainty quantification.", "date": "2026-05-12", "categories": []} {"id": "2605.12195", "title": "Fair Conformal Classification via Learning Representation-Based Groups", "abstract": "Conformal prediction methods provide statistically rigorous marginal coverage guarantees for machine learning models, but such guarantees fail to account for algorithmic biases, thereby undermining fairness and trust. This paper introduces a fair conformal inference framework for classification tasks. The proposed method constructs prediction sets that guarantee conditional coverage on adaptively identified subgroups, which can be implicitly defined through nonlinear feature combinations. By balancing effectiveness and efficiency in producing compact, informative prediction sets and ensuring adaptive equalized coverage across unfairly treated subgroups, our approach paves a practical pathway toward trustworthy machine learning. Extensive experiments on both synthetic and real-world datasets demonstrate the effectiveness of the framework.", "date": "2026-05-12", "categories": []} {"id": "2605.12341", "title": "Multi-Variable Conformal Prediction: Optimizing Prediction Sets without Data Splitting", "abstract": "Conformal prediction constructs prediction sets with finite-sample coverage guarantees, but its calibration stage is structurally constrained to a scalar score function and a single threshold variable - forcing shapes of prediction sets to be fixed before calibration, typically through data splitting. We introduce multi-variable conformal prediction (MCP), a framework that extends conformal prediction to vector-valued score functions with multiple simultaneous calibration variables. Building on scenario theory as a principled framework for certifying data-driven decisions, MCP unifies prediction set design and calibration into a single optimization problem, eliminating data splitting without sacrificing coverage guarantees. We propose two computationally efficient variants: RemMCP, grounded in constrained optimization with constraint removal, which admits a clean generalization of split conformal prediction; and RelMCP, based on iterative optimization with constraint relaxation, which supports non-convex score functions at the cost of possibly greater conservatism. Through numerical experiments on ellipsoidal and multi-modal prediction sets, we demonstrate that RemMCP and RelMCP consistently meet the target coverage with prediction set sizes smaller than or comparable to those of baselines with data split, while considerably reducing variance across calibration runs - a direct consequence of using all available data for shape optimization and calibration simultaneously.", "date": "2026-05-12", "categories": []} {"id": "2605.08561", "title": "CONTRA: Conformal Prediction Region via Normalizing Flow Transformation", "abstract": "Density estimation and reliable prediction regions for outputs are crucial in supervised and unsupervised learning. While conformal prediction effectively generates coverage-guaranteed regions, it struggles with multi-dimensional outputs due to reliance on one-dimensional nonconformity scores. To address this, we introduce CONTRA: CONformal prediction region via normalizing flow TRAnsformation. CONTRA utilizes the latent spaces of normalizing flows to define nonconformity scores based on distances from the center. This allows for the mapping of high-density regions in latent space to sharp prediction regions in the output space, surpassing traditional hyperrectangular or elliptical conformal regions. Further, for scenarios where other predictive models are favored over flow-based models, we extend CONTRA to enhance any such model with a reliable prediction region by training a simple normalizing flow on the residuals. We demonstrate that both CONTRA and its extension maintain guaranteed coverage probability and outperform existing methods in generating accurate prediction regions across various datasets. We conclude that CONTRA is an effective tool for (conditional) density estimation, addressing the under-explored challenge of delivering multi-dimensional prediction regions.", "date": "2026-05-08", "categories": []} {"id": "2505.14113", "title": "CONSIGN: Conformal Segmentation Informed by Spatial Groupings via Decomposition", "abstract": "Most machine learning-based image segmentation models produce pixel-wise confidence scores that represent the model's predicted probability for each class label at every pixel. While this information can be particularly valuable in high-stakes domains such as medical imaging, these scores are heuristic in nature and do not constitute rigorous quantitative uncertainty estimates. Conformal prediction (CP) provides a principled framework for transforming heuristic confidence scores into statistically valid uncertainty estimates. However, applying CP directly to image segmentation ignores the spatial correlations between pixels, a fundamental characteristic of image data. This can result in overly conservative and less interpretable uncertainty estimates. To address this, we propose CONSIGN (Conformal Segmentation Informed by Spatial Groupings via Decomposition), a CP-based method that incorporates spatial correlations to improve uncertainty quantification in image segmentation. Our method generates meaningful prediction sets that come with user-specified, high-probability error guarantees. It is compatible with any pre-trained segmentation model capable of generating multiple sample outputs. We evaluate CONSIGN against two CP baselines across three medical imaging datasets and two COCO dataset subsets, using three different pre-trained segmentation models. Results demonstrate that accounting for spatial structure significantly improves performance across multiple metrics and enhances the quality of uncertainty estimates.", "date": "2025-05-20", "categories": []} {"id": "2602.03331", "title": "Bayesian Conformal Prediction as a Decision Risk Problem", "abstract": "We propose Bayesian Conformal Prediction (BCP), a framework that combines Bayesian posterior predictive distributions with PAC-style conformal risk control to produce prediction sets with finite-sample coverage guarantees. Standard quantile-threshold conformal methods often construct prediction sets using a single fixed threshold, which typically yields connected prediction sets. While valid, such sets can be inefficient when the posterior predictive distribution is multimodal, since they may span low-density regions between separated modes. The main contribution of BCP is to formulate conformal prediction as a decision-risk optimisation problem, extending standard fixed quantile-threshold sets to optimised highest posterior density (HPD) prediction sets. These sets can be disjoint, concentrating probability mass on separated high-density regions. Validity is enforced using a PAC-style risk constraint, which provides coverage control even when the Bayesian model is misspecified. In standard nested-threshold settings, BCP recovers the smallest feasible threshold, aligning with existing PAC-based approaches. In the multimodal experiment, HPD geometry substantially improves efficiency, reducing mean prediction set size from $4.82$ to $2.07$ while satisfying the target PAC pass rate. Across regression, classification, and distribution-shift experiments, BCP maintains reliable coverage under model misspecification, whereas Bayesian credible intervals can fail to preserve nominal coverage.", "date": "2026-02-03", "categories": []} {"id": "2605.05497", "title": "Online Localized Conformal Prediction", "abstract": "Conformal prediction is a framework that provides valid uncertainty quantification for general models with exchangeable data. However, in the online learning and time-series settings, exchangeability is not satisfied. Existing online conformal methods, such as adaptive conformal inference (ACI), can achieve long-run validity, yet they remain inefficient under covariate heterogeneity because they rely on global calibration. We propose \\emph{Online Localized Conformal Prediction (OLCP)}, which combines online adaptation with covariate-dependent localization to better reflect heterogeneity. To reduce sensitivity to the localization bandwidth, we further develop \\emph{OLCP-Hedge}, which performs bandwidth selection as an online expert aggregation problem using a constrained online convex optimization framework. Importantly, we provide coverage guarantees for both algorithms and demonstrate through simulations and real-data experiments that the proposed methods attain valid long-run coverage with narrower prediction sets than existing baselines.", "date": "2026-05-06", "categories": []} {"id": "2605.07100", "title": "TRACE: Transport Alignment Conformal Prediction via Diffusion and Flow Matching Models", "abstract": "Constructing valid and informative conformal prediction regions for multi-dimensional outputs remains a fundamental challenge. While conformal prediction provides finite-sample, distribution-free coverage guarantees, its practical performance critically depends on the choice of nonconformity score. Existing approaches often rely on restrictive geometric assumptions or require explicit likelihood evaluation and invertible transformations, limiting their applicability in complex generative settings.\n In this work, we introduce TRACE (TRansport Alignment Conformal Estimation), a conformal prediction framework that defines nonconformity through transport alignment in diffusion and flow matching models. Rather than evaluating likelihoods, we measure how well a candidate output aligns with the learned generative dynamics by averaging denoising or velocity-matching errors along stochastic transport trajectories. The resulting transport-based scores are scalar-valued and can be calibrated using split conformal prediction, yielding valid marginal coverage under exchangeability. We further analyze the statistical properties of the proposed scores and their sensitivity to computational budget. Experiments on synthetic and real datasets demonstrate valid coverage and show that the resulting regions adapt naturally to multimodal and non-convex conditional distributions.", "date": "2026-05-08", "categories": []} {"id": "2605.07115", "title": "Conformal-Style Quantile Analyses for Stochastic Bandits", "abstract": "Stochastic bandit algorithms are usually analyzed under a mean-reward criterion, yet many problems favor arms with strong upper-tail performance, which we study herein. For a fixed miscoverage level \\(\\alpha\\), the natural upper-tail target of arm \\(j\\) is the upper endpoint \\(F_j^{-1}(1-\\alpha/2)\\) of a central prediction interval. This target can rank arms differently from their means, creating a central mismatch with the classical bandit objective. To this end, we propose ACP-UCB1, a conformal-style policy that combines an adaptive conformal estimate of the upper endpoint with a UCB-type optimism bonus. The technical challenge is that the conformity scores used by ACP-UCB1 are recomputed from evolving empirical quantile estimates and evaluated at an adaptive level. We control this endpoint through reward-quantile concentration, a perturbation argument for recomputed score quantiles, and deterministic localization of the adaptive level. ACP-UCB1 achieves logarithmic upper-quantile regret with per-arm contribution \\(O(\\nicefrac{\\log n}{\\Delta_j^{\\mathrm{ACP}}})\\). We also provide metric-specific regret decompositions comparing ACP-UCB1 with UCB1 and use numerical experiments to validate performance and improvement.", "date": "2026-05-08", "categories": []} {"id": "2605.07549", "title": "Probabilistic Object Detection with Conformal Prediction", "abstract": "Conformal Prediction (CP) is a distribution-free method for constructing prediction sets with marginal finite-sample coverage guarantees, making it a suitable framework for reliable uncertainty quantification in safety-critical object detection. However, object detection introduces structured multi-output predictions, complicating the application of classical CP theory developed for single outputs. In addition, standard, unscaled CP produces fixed-width prediction intervals across inputs, leading to unnecessary width for low-uncertainty predictions. While scaled CP addresses this by adapting the interval width to an input-dependent uncertainty estimate, prior work has neither systematically compared unscaled and scaled CP for multi-class object detection, nor integrated CP with a complementary uncertainty quantification method in this setting. We fill this gap by: (i) applying CP coordinate-wise to bounding box corners with a Bonferroni correction for box-level guarantees; (ii) scaling the resulting intervals using per-prediction aleatoric uncertainty estimates derived from a probabilistic object detector trained with loss attenuation, evaluated in uncalibrated and two calibrated variants; (iii) extending to a two-step pipeline that constructs prediction sets for the class using RAPS and conditions the conformalized bounding boxes on the predicted class set. Across three autonomous driving datasets (KITTI, BDD, CODA), including a cross-domain setting under distribution shift, scaled CP consistently improves interval sharpness over unscaled CP, achieving up to 19% higher IoU and 39% lower interval scores, without sacrificing coverage. Class-wise calibration further improves coverage for both variants with a negligible effect on sharpness. Together, these improvements yield more actionable uncertainty estimates for real-time, real-world object detection.", "date": "2026-05-08", "categories": []} {"id": "2605.07963", "title": "Aggregation in conformal e-classification", "abstract": "Aggregating conformal predictors is a standard way of balancing their predictive and computational efficiency while retaining their validity, at least approximately. An important advantage of conformal e-predictors is that they are easier to aggregate without sacrificing their validity. This paper studies experimentally cross-conformal e-prediction, which is an existing method of aggregating conformal e-predictors, and its modifications that are conceptually simpler and more flexible.", "date": "2026-05-08", "categories": []} {"id": "2605.08074", "title": "GRAPHLCP: Structure-Aware Localized Conformal Prediction on Graphs", "abstract": "Conformal prediction (CP) provides a distribution-free approach to uncertainty quantification with finite-sample guarantees. However, applying CP to graph neural networks (GNNs) remains challenging as the combinatorial nature of graphs often leads to insufficiently certain predictions and indiscriminative embeddings. Existing methods primarily rely on embedding-space proximity for localization, which can be unreliable for graphs and yield inefficient prediction sets. We propose GRAPHLCP, a proximity-based localized CP framework that explicitly incorporates graph topology and inter-node dependencies into localization and weighting. Our approach introduces a feature-aware densification step to mitigate locality bias in sparse graphs, followed by a Personalized PageRank-based kernel computation to model structural proximity. This enables topology-dependent anchor sampling and calibration weighting that captures both local and long-range dependencies. Extensive experiments on several regression and classification datasets demonstrate that GRAPHLCP guarantees marginal coverage with finite samples while efficiently attaining favorable test conditional coverage across various conditioning scenarios.", "date": "2026-05-08", "categories": []} {"id": "2507.20941", "title": "Multivariate Standardized Residuals for Conformal Prediction", "abstract": "While split conformal prediction guarantees marginal coverage, approaching the stronger property of conditional coverage is essential for reliable uncertainty quantification. Naive conformal scores, however, suffer from poor conditional coverage in heteroskedastic settings. In univariate regression, this is commonly addressed by normalizing non-conformity scores using an estimated local score variance. In this work, we propose a natural extension of this normalization to the multivariate setting, effectively whitening the residuals to decouple output correlations and standardize local variance. Furthermore, we derive a sufficient condition characterizing a broad class of distributions for which standardized residuals yield asymptotic conditional coverage. We demonstrate that using the Mahalanobis distance induced by a learned local covariance as a non-conformity score provides a closed-form, computationally efficient mechanism for capturing inter-output correlations and heteroskedasticity, avoiding the expensive sampling required by previous methods based on cumulative distribution functions. This structure unlocks several practical extensions, including the handling of missing output values, the refinement of conformal sets when partial information is revealed, and the construction of valid conformal sets for transformations of the output. Finally, we provide extensive empirical evidence on both synthetic and real-world datasets showing that our approach yields conformal sets that improve upon the conditional coverage of existing multivariate baselines.", "date": "2025-07-28", "categories": []} {"id": "2602.07633", "title": "Flow-Based Conformal Predictive Distributions", "abstract": "Conformal prediction provides a distribution-free framework for uncertainty quantification via prediction sets with exact finite-sample coverage. In low dimensions these sets are easy to interpret, but in high-dimensional or structured output spaces they are difficult to represent and use, which can limit their ability to integrate with downstream tasks such as sampling and probabilistic forecasting. We show that any sufficiently regular differentiable nonconformity score induces a deterministic flow on the output space whose trajectories converge to the boundary of the corresponding conformal prediction set. This leads to a computationally efficient, training-free method for sampling conformal boundaries in arbitrary dimensions. Mixing across confidence levels yields conformal predictive distributions whose quantile regions coincide with the empirical conformal prediction sets. We provide an approximation bound decomposing CPD predictive error into score-induced distortion, base-measure quality, and gradient flow-induced distortion. We evaluate the approach on PDE inverse problems, precipitation downscaling, climate model debiasing, and hurricane trajectory forecasting.", "date": "2026-02-07", "categories": []} {"id": "2605.06204", "title": "When Does Trimming Help Conformal Prediction? A Retained-Law Diagnostic under Calibration Contamination", "abstract": "Trimming suspicious calibration points is a common response to contamination in conformal prediction. Its effect on clean-target coverage, however, is governed by the retained law induced by trimming, not by the contamination level alone. We analyse fixed-threshold trimming as conditioning rather than purification. It replaces the contaminated calibration law with a retained law, reducing clean-target coverage to a one-dimensional score-CDF transfer problem with an exact finite-sample identity. A componentwise bound on the transfer gap gives a population-level diagnostic. This separates a clean-side covariance cost from a retained-contamination cost, governed by the dirty-to-clean retention ratio. Trimming helps when the anomaly score separates retention probabilities while remaining score-neutral on the clean population. Otherwise, it cannot substantially reduce contamination through the retained mixture coefficient. We also give finite-sample certificate templates that provide numerical guarantees under independent audit.", "date": "2026-05-07", "categories": []} {"id": "2605.06646", "title": "Inductive Venn-Abers and related regressors", "abstract": "Venn-Abers predictors are probabilistic predictors that enjoy appealing properties of validity, but their major limitation is that they are applicable only to the case of binary classification, with a recent extension to bounded regression. We generalize them to the case of unbounded regression, which requires adding an element of conformal prediction. In our simulation and empirical studies we investigate the predictive efficiency of point regressors derived from Venn-Abers regressors and argue that they somewhat improve the predictive efficiency of standard regressors for larger training sets.", "date": "2026-05-07", "categories": []} {"id": "2605.04957", "title": "Delving into Non-Exchangeability for Conformal Prediction in Graph-Structured Multivariate Time Series", "abstract": "Point forecasting for graph-structured multivariate time series is a fundamental problem, but rigorous uncertainty quantification for such predictions is still underexplored. Conformal prediction (CP) offers uncertainty estimation with a solid coverage guarantee under the exchangeability assumption, which requires the joint data distribution to be unchanged under permutation. However, in graph-structured time series, inherent cross-node coupling can violate the exchangeability condition, making direct application of CP unreliable. Inspired by the spectral graph theory, such coupling resides in global trends and can be characterized by the low-frequency components, while high-frequency components are nearly exchangeable. Therefore, we propose a novel concept named Spectral Graph Conditional Exchangeability (SGCE), which conditions exchangeable high-frequency components on low-frequency ones to preserve global trends and enable effective CP in the spectral domain. Based on SGCE, we further propose Spectral Conformal prediction via wAveLEt transform (SCALE). SCALE uses graph wavelets to decompose low/high-frequency components and conformalizes high-frequency residuals via adaptive gating over a low-frequency embedding. Experimental results on real-world traffic datasets show that SCALE not only achieves valid coverage but also consistently improves the coverage-efficiency trade-off over the state-of-the-art CP methods.", "date": "2026-05-06", "categories": []} {"id": "2512.01384", "title": "CLAPS: Aleatoric-Epistemic Scaling via Last-Layer Laplace for Conformal Regression", "abstract": "Conformal regression provides finite-sample marginal coverage, but it does not by itself determine how interval width should adapt across heterogeneous inputs. Existing locally adaptive methods mainly account for aleatoric noise, leaving uncertainty from weak training support less explicit. We propose Conformal Laplace-Aware Predictive Scaling (CLAPS), a split conformal regression method that uses heteroscedastic last-layer Laplace uncertainty as the local normalization scale. CLAPS combines learned input-dependent noise with last-layer epistemic uncertainty, while retaining validity through standard conformal calibration. We characterize this aleatoric--epistemic scale, derive its heteroscedastic last-layer precision, and show that it reduces to aleatoric local scaling as epistemic uncertainty contracts. Experiments show nominal-level coverage with competitive interval efficiency.", "date": "2025-12-01", "categories": []} {"id": "2605.03233", "title": "Conformalized Percentile Interval: Finite Sample Validity and Improved Conditional Performance", "abstract": "Conformal prediction provides distribution-free predictive intervals with finite-sample marginal coverage. However, achieving conditional validity and interval efficiency (in terms of short interval length) remains challenging, particularly in complex settings with heteroskedasticity, skewed responses, or estimation errors. We propose a conformal-style calibration method for responses obtained by the probability integral transform (PIT) of the conditional cumulative distribution function (CDF) estimated via neural networks to construct a finite-sample-adjusted percentile interval with the shortest length determined by the estimated conditional CDF. Calibrating in PIT space is effective because PIT values are asymptotically feature-independent when the CDF estimator is accurate, which mitigates feature-dependent miscoverage and improves conditional calibration. On the other hand, our percentile calibration adapts to the empirical PIT distribution, which is robust against a possibly imperfect estimation of the conditional CDF. We prove the finite-sample marginal coverage property of the proposed method and show its asymptotic conditional coverage under mild consistency conditions. Experiments on diverse synthetic and real-world benchmarks demonstrate better conditional calibration and substantially shorter intervals than existing methods.", "date": "2026-05-04", "categories": []} {"id": "2507.04441", "title": "A Category-Theoretic Analysis of Conformal Prediction", "abstract": "Conformal prediction (CP) produces prediction regions with finite-sample, distribution free coverage guarantees, but its interpretation as a quantitative uncertainty tool is often left implicit. We develop a category-theoretic approach that makes this structure explicit. We show that Full Conformal Prediction can be represented as a morphism in two categories capturing (i) stability of set-valued procedures and (ii) measurability of random regions. Under mild conditions, we prove a commuting diagram result that decomposes the construction of a conformal region into two steps: Extracting a set of predictive distributions from the data, and then deriving a prediction region from this set. This decomposition provides a principled route to numerical uncertainty summaries beyond region size. We further prove an asymptotic compatibility result showing that, for Bayesian predictive scores in regular regimes, conformal regions converge to Bayesian predictive density level sets; We also provide quantitative rates under local empirical process and boundary regularity assumptions. This highlights a bridge between Bayesian, frequentist, and imprecise probabilistic prediction. We additionally identify conditions under which upper posterior constructions are related to e-posteriors, clarifying when e-value-based and conformal-imprecise representations can coincide. Finally, we show that the region extractor is functorial; This yields a modular privacy-compatible perspective in which privacy-preserving outer approximations of shared summary objects lead to conservative global prediction regions.", "date": "2025-07-06", "categories": []} {"id": "2605.01452", "title": "Stable Localized Conformal Prediction via Transduction", "abstract": "Existing evaluations of conformal prediction, such as prediction efficiency and test-conditional coverage, are defined in expectation over the calibration data. In practice, when only one calibration set of limited size is available, prediction sets often exhibit high variability in size, especially for methods with localization. We formalize this concern as set stability, defined as the variance of the conditional expectation of the set size given the calibration data. To improve stability without requiring additional target-task labels, we propose Stable Conformal Prediction (StCP), a transfer learning approach that utilizes labeled source-task data and unlabeled target data. Theoretically, we characterize the marginal coverage and stability of StCP; empirically, it delivers more stable prediction sets than standard conformal prediction methods, especially for those with localization, when calibration data are limited.", "date": "2026-05-02", "categories": []} {"id": "2605.01868", "title": "Robust Conditional Conformal Prediction via Branched Normalizing Flow", "abstract": "Conformal prediction (CP) constructs prediction sets with marginal coverage guarantees under the assumption that the calibration and test distributions are identical. However, under distribution shift, existing approaches primarily align marginal conformal score distributions, which is sufficient to preserve marginal coverage but does not control the conditional coverage error at individual test inputs. As a consequence, CP can remain unreliable in regions where the conditional score distributions are mismatched. In this work, we bound the conditional invalidity of CP under distribution shift in terms of the Wasserstein distance between the calibration and test distributions. This result highlights the role of invertible transport in mitigating conditional coverage degradation. Motivated by this insight, we introduce Branched Normalizing Flow (BNF), a two-branch architecture that normalizes a test input to the calibration distribution and transforms the prediction set of the normalized input back to the test distribution while preserving conditional guarantees. Empirically, BNF consistently improves conditional coverage robustness on nine datasets across a wide range of confidence levels.", "date": "2026-05-03", "categories": []} {"id": "2605.02072", "title": "Weight Clipping for Robust Conformal Inference under Unbounded Covariate Shifts", "abstract": "Conformal prediction (CP) provides powerful, distribution-free prediction sets, but its guarantees rely on the exchangeability of training and test data, which is often violated in practice due to covariate shifts. While weighted conformal prediction (WCP) is designed to handle such shifts, it can suffer from significant undercoverage when the density ratio between the distributions is unbounded and/or must be learned. This is because of both overfitting in learning the density ratio, and high variance in estimating the nonconformity score threshold. To address this, we introduce clipped least-squares importance fitting (CLISF) as a reduced-variance method for density ratio estimation. Specifically, we show that density ratios learned using CLISF, when plugged into WCP, have bounded expected undercoverage. Furthermore, we show that the undercoverage can be corrected by running WCP with a slightly inflated coverage target; crucially, we are able to estimate the required level of inflation from the data. We provide the first theoretical guarantees for weight clipping in conformal inference, achieving dataset-conditional coverage with a sample complexity that does not blow up with the higher moments of the true density ratio -- a key limitation of prior work. We verify our results on real-world benchmarks and synthetic data.", "date": "2026-05-03", "categories": []} {"id": "2605.00126", "title": "SPLICE: Latent Diffusion over JEPA Embeddings for Conformal Time-Series Inpainting", "abstract": "Generative models for time-series imputation achieve strong reconstruction accuracy, yet provide no finite-sample reliability guarantees, a critical limitation in power systems where imputed values inform dispatch and planning. We introduce SPLICE (Self-supervised Predictive Latent Inpainting with Conformal Envelopes), a modular framework coupling latent generative imputation with distribution-free, online-adaptive prediction intervals. A JEPA encoder maps daily load segments into a 64-dimensional latent space; a conditional latent bridge with four sampling modes generates candidate gap trajectories; an hourly-conditioned decoder maps back to signal space; and Adaptive Conformal Inference (ACI) wraps the output with coverage-guaranteed prediction bands. The flow-matching variant achieves comparable quality to DDIM in 5--10 ODE steps (5-10x speedup). On thirteen load datasets (nine proprietary, three UCI Electricity, ETTh1), SPLICE achieves the lowest mean Load-only MSE (0.056), winning 9/12 non-degenerate datasets at 91-day gaps and 18/32 across all gap lengths vs. five established baselines, and produces the best CRPS (0.161, -18.3% vs. the strongest competitor). ACI delivers 93--95% empirical coverage, correcting under-coverage failures of up to 7.5 pp observed with static conformal prediction. A pooled JEPA encoder trained on nine feeds transfers to four unseen domains, matching or exceeding per-dataset oracles with only a quick bridge fine-tuning.", "date": "2026-04-30", "categories": []} {"id": "2605.00233", "title": "Adaptive Geodesic Conformal Prediction for Egocentric Camera Pose Estimation", "abstract": "Egocentric pose estimation for Augmented Reality (AR) and assistive devices requires not just accurate predictions but guaranteed uncertainty regions. Conformal prediction (CP) provides such guarantees without retraining, but we show that standard CP with a single fixed threshold achieves nominal 90% overall coverage while covering only ~60% of the hardest 25% of frames (Q4) -- a ~30 percentage-point conditional coverage gap consistent across 12 participants, 3 predictors, and 3 horizons (108 evaluations) on EPIC-Fields. We further show that a geodesic SE(3) nonconformity score identifies physically harder frames than Euclidean scoring, with only 15-26% Q4 overlap and 2-3x higher ground-truth camera displacement for geodesic Q4 frames. To close the coverage gap, we propose DINOv2-Bridge adaptive CP: a two-stage difficulty estimator trained on a single source participant that transfers cross-participant without any images at test time, improving Q4 coverage from ~0.75 to ~0.93 while maintaining overall coverage at the 90% target.", "date": "2026-04-30", "categories": []} {"id": "2602.19483", "title": "Making Conformal Predictors Robust in Healthcare Settings: a Case Study on EEG Classification", "abstract": "Quantifying uncertainty in clinical predictions is critical for high-stakes diagnosis tasks. Conformal prediction offers a principled approach by providing prediction sets with theoretical coverage guarantees. However, in practice, patient distribution shifts violate the i.i.d. assumptions underlying standard conformal methods, leading to poor coverage in healthcare settings. In this work, we evaluate several conformal prediction approaches on EEG seizure classification, a task with known distribution shift challenges and label uncertainty. We demonstrate that personalized calibration strategies can improve coverage by over 20 percentage points while maintaining comparable prediction set sizes. Our implementation is available via PyHealth, an open-source healthcare AI framework: https://github.com/sunlabuiuc/PyHealth.", "date": "2026-02-23", "categories": []} {"id": "2604.27149", "title": "ConformaDecompose: Explaining Uncertainty via Calibration Localization", "abstract": "Conformal Prediction provides distribution-free prediction intervals with guaranteed coverage, but its reliance on a single global calibration threshold obscures the sources of uncertainty at the instance level. In particular, it conflates irreducible noise with uncertainty induced by heterogeneous training data (aleatoric), model limitations, or calibration mismatch (epistemic), offering little insight into why an interval is wide or whether it could be reduced. We introduce an uncertainty-aware explainability framework that analyses the reducibility of calibration-induced epistemic conformal uncertainty via progressive calibration localisation for regression tasks. The approach is diagnostic rather than causal: it does not estimate true aleatoric or epistemic uncertainty, but explains how conformal intervals contract and stabilise as calibration support is localised around a test instance. Across benchmarks and real-world data, absolute reducible uncertainty aligns with epistemic proxies, while its relative contribution varies by task, revealing regimes hidden by interval width. This instance-level view complements conformal uncertainty, enhancing interpretability without altering the predictor or coverage.", "date": "2026-04-29", "categories": []} {"id": "2604.27914", "title": "Geometry-Calibrated Conformal Abstention for Language Models", "abstract": "When language models lack relevant knowledge for a given query, they frequently generate plausible responses that can be hallucinations, rather than admitting being agnostic about the answer. Retraining models to reward admitting ignorance can lead to overly conservative behaviors and poor generalization due to scarce evaluation benchmarks. We propose a post hoc framework, Conformal Abstention (CA), adapted from conformal prediction (CP) to determine whether to abstain from answering a query. CA provides finite-sample guarantees on both the probability of participation (i.e., not abstaining) and the probability that the generated response is correct. Importantly, the abstention decision relies on prediction confidence rather than the non-conformity scores used in CP, which are intractable for open-ended generation. To better align prediction confidence with the model's ignorance, we introduce a calibration strategy using representation geometry within the model to measure knowledge involvement in shaping the response. Experiments demonstrate that we improve selective answering significantly with 75 percent conditional correctness.", "date": "2026-04-30", "categories": []} {"id": "2512.12844", "title": "Selective Conformal Risk Control", "abstract": "Reliable uncertainty quantification is essential for deploying machine learning systems in high-stakes domains. Conformal prediction provides distribution-free coverage guarantees but often produces overly large prediction sets, limiting its practical utility. To address this issue, we propose \\textit{Selective Conformal Risk Control} (SCRC), a unified framework that integrates conformal prediction with selective classification. The framework formulates uncertainty control as a two-stage problem: the first stage selects confident samples for prediction, and the second stage applies conformal risk control on the selected subset to construct calibrated prediction sets. We develop two algorithms under this framework. The first, SCRC-T, preserves exchangeability by computing thresholds jointly over calibration and test samples, offering exact finite-sample guarantees. The second, SCRC-I, is a calibration-only variant that provides PAC-style probabilistic guarantees while being more computational efficient. Experiments on two public datasets show that both methods achieve the target coverage and risk levels, with nearly identical performance, while SCRC-I exhibits slightly more conservative risk control but superior computational practicality. Our results demonstrate that selective conformal risk control offers an effective and efficient path toward compact, reliable uncertainty quantification.", "date": "2025-12-14", "categories": []} {"id": "2604.22391", "title": "Conformalized Super Learner", "abstract": "The Super Learner (SL) is a widely used ensemble method that combines predictions from a library of learners based on their predictive performance. Interval predictions are of considerable practical interest because they allow uncertainty in predictions produced by an individual learner or an ensemble to be quantified. Several methods have been proposed for constructing interval predictions based on the SL, however, these approaches are typically justified using asymptotic arguments or rely on computationally intensive procedures such as the bootstrap. Conformal prediction (CP) is a machine learning framework for constructing prediction intervals with finite-sample and asymptotic coverage guarantees under mild conditions. We propose coupling CP with the SL through a natural construction that mirrors the original SL framework, using individual learner weights and combining learner-specific conformity scores via a weighted majority vote. We characterize the properties of the resulting SL-based prediction intervals for continuous outcomes. We cover settings under exchangeability, potential violations of exchangeability, and data-generating mechanisms exhibiting heteroscedasticity, sparsity, and other forms of distributional heterogeneity. A comprehensive simulation study shows that the conformalized SL achieves valid finite-sample coverage with competitive performance relative to the true data-generating mechanism. A central contribution of this work is an application to predicting creatinine levels using socio-demographic, biometric, and laboratory measurements. This example demonstrates the benefits of an ensemble with carefully selected learners designed to capture key aspects of complex regression functions, including non-linear effects, interactions, sparsity, heteroscedasticity, and robustness to outliers.R", "date": "2026-04-24", "categories": []} {"id": "2603.27189", "title": "Conformal Prediction Assessment: A Framework for Conditional Coverage Evaluation and Selection", "abstract": "Conformal prediction provides rigorous distribution-free finite-sample guarantees for marginal coverage under the assumption of exchangeability, but may exhibit systematic undercoverage or overcoverage for specific subpopulations. Assessing conditional validity is challenging, as standard stratification methods suffer from the curse of dimensionality. We propose Conformal Prediction Assessment (CPA), a framework that reframes the evaluation of conditional coverage as a supervised learning task by training a reliability estimator that predicts instance-level coverage probabilities. Building on this estimator, we introduce the Conditional Validity Index (CVI), which decomposes reliability into safety (undercoverage risk) and efficiency (overcoverage cost). We establish convergence rates for the reliability estimator and prove the consistency of CVI-based model selection. Extensive experiments on synthetic and real-world datasets demonstrate that CPA effectively diagnoses local failure modes and that CC-Select, our CVI-based model selection algorithm, consistently identifies predictors with superior conditional coverage performance.", "date": "2026-03-28", "categories": []} {"id": "2604.21595", "title": "A Kernel Nonconformity Score for Multivariate Conformal Prediction", "abstract": "Multivariate conformal prediction requires nonconformity scores that compress residual vectors into scalars while preserving certain implicit geometric structure of the residual distribution. We introduce a Multivariate Kernel Score (MKS) that produces prediction regions that explicitly adapt to this geometry. We show that the proposed score resembles the Gaussian process posterior variance, unifying Bayesian uncertainty quantification with the coverage guarantees of frequentist-type. Moreover, the MKS can be decomposed into an anisotropic Maximum Mean Discrepancy (MMD) that interpolates between kernel density estimation and covariance-weighted distance. We prove finite-sample coverage guarantees and establish convergence rates that depend on the effective rank of the kernel-based covariance operator rather than the ambient dimension, enabling dimension-free adaptation. On regression tasks, the MKS reduces the volume of prediction region significantly, compared to ellipsoidal baselines while maintaining nominal coverage, with larger gains at higher dimensions and tighter coverage levels.", "date": "2026-04-23", "categories": []} {"id": "2604.20098", "title": "Differentiable Conformal Training for LLM Reasoning Factuality", "abstract": "Large Language Models (LLMs) frequently hallucinate, limiting their reliability in critical applications. Conformal Prediction (CP) addresses this by calibrating error rates on held-out data to provide statistically valid confidence guarantees. Recent work extends CP to LLM factuality to filter out risky claims, ensuring that hallucination rates remain below a user-specified level (e.g., 10%). While prior methods treat claims independently, Coherent Factuality extends to multi-step reasoning by representing outputs as dependency graphs and jointly validating claims with their logical ancestors. A key limitation is that Coherent Factuality is not differentiable, requiring hand-crafted scorers that at high reliability levels remove nearly 60% of true claims. We introduce Differentiable Coherent Factuality (DCF), a fully differentiable relaxation that enables learning improved scorers while provably recovering the original algorithm's guarantees. Experiments on two benchmark reasoning datasets demonstrate DCF achieves up to 141% improvement in claim retention while maintaining reliability guarantees, representing a significant step towards reliable conformal LLM systems.", "date": "2026-04-22", "categories": []} {"id": "2602.03986", "title": "eCP: Equivariant Conformal Prediction with pre-trained models", "abstract": "Conformal prediction, a post-hoc, distribution-free, finite-sample method of uncertainty quantification that offers formal coverage guarantees under the assumption of data exchangeability. Unfortunately, the resulting uncertainty regions can grow significantly in long horizon missions, rendering the statistical guarantees uninformative. To that end, we propose infusing CP with geometric information via group-averaging of the pretrained predictor to distribute the non-conformity mass across the orbits. Each sample now is treated as a representative of an orbit, thus uncertainty can be mitigated by other samples entangled to it via the orbit inducing elements of the symmetry group. Our approach provably yields contracted non-conformity scores in increasing convex order, implying improved exponential-tail bounds and sharper conformal prediction sets in expectation, especially at high confidence levels. We then propose an experimental design to test these theoretical claims in pedestrian trajectory prediction.", "date": "2026-02-03", "categories": []} {"id": "2604.01502", "title": "Conformal Risk Control under Non-Monotone Losses: Theory and Finite-Sample Guarantees", "abstract": "Conformal risk control (CRC) provides distribution-free guarantees for controlling the expected loss at a user-specified level. Existing theory typically assumes that the loss decreases monotonically with a tuning parameter that governs the size of the prediction set. However, this assumption is often violated in practice, where losses may behave non-monotonically due to competing objectives such as coverage and efficiency.\n In this paper, we study CRC under non-monotone loss functions when the tuning parameter is selected from a finite grid, a setting commonly arising in thresholding and discretized decision rules. Revisiting a known counterexample, we show that the validity of CRC without monotonicity depends critically on the relationship between the calibration sample size and the grid resolution. In particular, reliable risk control can still be achieved when the calibration sample is sufficiently large relative to the grid size.\n We establish a finite-sample guarantee for bounded losses over a grid of size $m$, showing that the excess risk above the target level $\\alpha$ scales on the order of $\\sqrt{\\log(m)/n}$, where $n$ is the calibration sample size. A matching lower bound demonstrates that this rate is minimax optimal. We also derive refined guarantees under additional structural conditions, including Lipschitz continuity and monotonicity, and extend the analysis to settings with distribution shift via importance weighting.\n Numerical experiments on synthetic multilabel classification and real object detection data illustrate the practical implications of non-monotonicity. Methods that explicitly account for finite-sample uncertainty achieve more stable risk control than approaches based on monotonicity transformations, while maintaining competitive prediction set sizes.", "date": "2026-04-02", "categories": []} {"id": "2604.16677", "title": "ReconVLA: An Uncertainty-Guided and Failure-Aware Vision-Language-Action Framework for Robotic Control", "abstract": "Vision-language-action (VLA) models have emerged as generalist robotic controllers capable of mapping visual observations and natural language instructions to continuous action sequences. However, VLAs provide no calibrated measure of confidence in their action predictions, thus limiting their reliability in real-world settings where uncertainty and failures must be anticipated. To address this problem we introduce ReconVLA, a reliable conformal model that produces uncertainty-guided and failure-aware control signals. Concretely, our approach applies conformal prediction directly to the action token outputs of pretrained VLA policies, yielding calibrated uncertainty estimates that correlate with execution quality and task success. Furthermore, we extend conformal prediction to the robot state space to detect outliers or unsafe states before failures occur, providing a simple yet effective failure detection mechanism that complements the action-level uncertainty. We evaluate ReconVLA in both simulation and real robot experiments across diverse manipulation tasks. Our results show that conformalized action predictions consistently improve failure anticipation, reduce catastrophic errors, and provide a calibrated measure of confidence without retraining or modifying the underlying VLA.", "date": "2026-04-17", "categories": []} {"id": "2604.17984", "title": "Online Conformal Prediction with Adversarial Semi-bandit Feedback via Regret Minimization", "abstract": "Uncertainty quantification is crucial in safety-critical systems, where decisions must be made under uncertainty. In particular, we consider the problem of online uncertainty quantification, where data points arrive sequentially. Online conformal prediction is a principled online uncertainty quantification method that dynamically constructs a prediction set at each time step. While existing methods for online conformal prediction provide long-run coverage guarantees without any distributional assumptions, they typically assume a full feedback setting in which the true label is always observed. In this paper, we propose a novel learning method for online conformal prediction with partial feedback from an adaptive adversary-a more challenging setup where the true label is revealed only when it lies inside the constructed prediction set. Specifically, we formulate online conformal prediction as an adversarial bandit problem by treating each candidate prediction set as an arm. Building on an existing algorithm for adversarial bandits, our method achieves a long-run coverage guarantee by explicitly establishing its connection to the regret of the learner. Finally, we empirically demonstrate the effectiveness of our method in both independent and identically distributed (i.i.d.) and non-i.i.d. settings, showing that it successfully controls the miscoverage rate while maintaining a reasonable size of the prediction set.", "date": "2026-04-20", "categories": []} {"id": "2604.18441", "title": "Conformal Robust Set Estimation", "abstract": "Conformal prediction provides finite-sample, distribution-free coverage under exchangeability, but standard constructions may lack robustness in the presence of outliers or heavy tails. We propose a robust conformal method based on a non-conformity score defined as the half-mass radius around a point, equivalently the distance to its $(\\lfloor n/2\\rfloor+1)$-nearest neighbour.\n We show that the resulting conformal regions are marginally valid for any sample size and converge in probability to a robust population central set defined through a distance-to-a-measure functional. Under mild regularity conditions, we establish exponential concentration and tail bounds that quantify the deviation between the empirical conformal region and its population counterpart. These results provide a probabilistic justification for using robust geometric scores in conformal prediction, even for heavy-tailed or multi-modal distributions.", "date": "2026-04-20", "categories": []} {"id": "2604.18482", "title": "Safe Control using Learned Safety Filters and Adaptive Conformal Inference", "abstract": "Safety filters have been shown to be effective tools to ensure the safety of control systems with unsafe nominal policies. To address scalability challenges in traditional synthesis methods, learning-based approaches have been proposed for designing safety filters for systems with high-dimensional state and control spaces. However, the inevitable errors in the decisions of these models raise concerns about their reliability and the safety guarantees they offer. This paper presents Adaptive Conformal Filtering (ACoFi), a method that combines learned Hamilton-Jacobi reachability-based safety filters with adaptive conformal inference. Under ACoFi, the filter dynamically adjusts its switching criteria based on the observed errors in its predictions of the safety of actions. The range of possible safety values of the nominal policy's output is used to quantify uncertainty in safety assessment. The filter switches from the nominal policy to the learned safe one when that range suggests it might be unsafe. We show that ACoFi guarantees that the rate of incorrectly quantifying uncertainty in the predicted safety of the nominal policy is asymptotically upper bounded by a user-defined parameter. This gives a soft safety guarantee rather than a hard safety guarantee. We evaluate ACoFi in a Dubins car simulation and a Safety Gymnasium environment, empirically demonstrating that it significantly outperforms the baseline method that uses a fixed switching threshold by achieving higher learned safety values and fewer safety violations, especially in out-of-distribution scenarios.", "date": "2026-04-20", "categories": []} {"id": "2604.11305", "title": "Beyond Fixed False Discovery Rates: Post-Hoc Conformal Selection with E-Variables", "abstract": "Conformal selection (CS) uses calibration data to identify test inputs whose unobserved outcomes are likely to satisfy a pre-specified minimal quality requirement, while controlling the false discovery rate (FDR). Existing methods fix the target FDR level before observing data, which prevents the user from adapting the balance between number of selected test inputs and FDR to downstream needs and constraints based on the available data. For example, in genomics or neuroimaging, researchers often inspect the distribution of test statistics, and decide how aggressively to pursue candidates based on observed evidence strength and available follow-up resources. To address this limitation, we introduce {post-hoc CS} (PH-CS), which generates a path of candidate selection sets, each paired with a data-driven false discovery proportion (FDP) estimate. PH-CS lets the user select any operating point on this path by maximizing a user-specified utility, arbitrarily balancing selection size and FDR. Building on conformal e-variables and the e-Benjamini-Hochberg (e-BH) procedure, PH-CS is proved to provide a finite-sample post-hoc reliability guarantee whereby the ratio between estimated FDP level and true FDP is, on average, upper bounded by $1$, so that the average estimated FDP is, to first order, a valid upper bound on the true FDR. PH-CS is extended to control quality defined in terms of a general risk. Experiments on synthetic and real-world datasets demonstrate that, unlike CS, PH-CS can consistently satisfy user-imposed utility constraints while producing reliable FDP estimates and maintaining competitive FDR control.", "date": "2026-04-13", "categories": []} {"id": "2604.16217", "title": "Beyond Surface Statistics: Robust Conformal Prediction for LLMs via Internal Representations", "abstract": "Large language models are increasingly deployed in settings where reliability matters, yet output-level uncertainty signals such as token probabilities, entropy, and self-consistency can become brittle under calibration--deployment mismatch. Conformal prediction provides finite-sample validity under exchangeability, but its practical usefulness depends on the quality of the nonconformity score. We propose a conformal framework for LLM question answering that uses internal representations rather than output-facing statistics: specifically, we introduce Layer-Wise Information (LI) scores, which measure how conditioning on the input reshapes predictive entropy across model depth, and use them as nonconformity scores within a standard split conformal pipeline. Across closed-ended and open-domain QA benchmarks, with the clearest gains under cross-domain shift, our method achieves a better validity--efficiency trade-off than strong text-level baselines while maintaining competitive in-domain reliability at the same nominal risk level. These results suggest that internal representations can provide more informative conformal scores when surface-level uncertainty is unstable under distribution shift.", "date": "2026-04-17", "categories": []} {"id": "2604.13253", "title": "Bias-Corrected Adaptive Conformal Inference for Multi-Horizon Time Series Forecasting", "abstract": "Adaptive Conformal Inference (ACI) provides distribution-free prediction intervals with asymptotic coverage guarantees for time series under distribution shift. However, ACI only adapts the quantile threshold -- it cannot shift the interval center. When a base forecaster develops persistent bias after a regime change, ACI compensates by widening intervals symmetrically, producing unnecessarily conservative bands. We propose Bias-Corrected ACI (BC-ACI), which augments standard ACI with an online exponentially weighted moving average (EWM) estimate of forecast bias. BC-ACI corrects nonconformity scores before quantile computation and re-centers prediction intervals, addressing the root cause of miscalibration rather than its symptom. An adaptive dead-zone threshold suppresses corrections when estimated bias is indistinguishable from noise, ensuring no degradation on well-calibrated data. In controlled experiments across 688 runs spanning two base models, four synthetic regimes, and three real datasets, BC-ACI reduces Winkler interval scores by 13--17% under mean and compound distribution shifts (Wilcoxon p < 0.001) while maintaining equivalent performance on stationary data (ratio 1.002x). We provide finite-sample analysis showing that coverage guarantees degrade gracefully with bias estimation error.", "date": "2026-04-14", "categories": []} {"id": "2604.13395", "title": "Quantifying and Understanding Uncertainty in Large Reasoning Models", "abstract": "Large Reasoning Models (LRMs) have recently demonstrated significant improvements in complex reasoning. While quantifying generation uncertainty in LRMs is crucial, traditional methods are often insufficient because they do not provide finite-sample guarantees for reasoning-answer generation. Conformal prediction (CP) stands out as a distribution-free and model-agnostic methodology that constructs statistically rigorous uncertainty sets. However, existing CP methods ignore the logical connection between the reasoning trace and the final answer. Additionally, prior studies fail to interpret the origins of uncertainty coverage for LRMs as they typically overlook the specific training factors driving valid reasoning. Notably, it is challenging to disentangle reasoning quality from answer correctness when quantifying uncertainty, while simultaneously establishing theoretical guarantees for computationally efficient explanation methods. To address these challenges, we first propose a novel methodology that quantifies uncertainty in the reasoning-answer structure with statistical guarantees. Subsequently, we develop a unified example-to-step explanation framework using Shapley values that identifies a provably sufficient subset of training examples and their key reasoning steps to preserve the guarantees. We also provide theoretical analyses of our proposed methods. Extensive experiments on challenging reasoning datasets verify the effectiveness of the proposed methods.", "date": "2026-04-15", "categories": []} {"id": "2604.13991", "title": "Adaptive Conformal Prediction for Improving Factuality of Generations by Large Language Models", "abstract": "Large language models (LLMs) are prone to generating factually incorrect outputs. Recent work has applied conformal prediction to provide uncertainty estimates and statistical guarantees for the factuality of LLM generations. However, existing approaches are typically not prompt-adaptive, limiting their ability to capture input-dependent variability. As a result, they may filter out too few items (leading to over-coverage) or too many (under-coverage) for a given task or prompt. We propose an adaptive conformal prediction approach that extends conformal score transformation methods to LLMs, with applications to long-form generation and multiple-choice question answering. This enables prompt-dependent calibration, retaining marginal coverage guarantees while improving conditional coverage. In addition, the approach naturally supports selective prediction, allowing unreliable claims or answer choices to be filtered out in downstream applications. We evaluate our approach on multiple white-box models across diverse domains and show that it significantly outperforms existing baselines in terms of conditional coverage.", "date": "2026-04-15", "categories": []} {"id": "2501.19038", "title": "Conformal Prediction in Hierarchical Classification with Constrained Representation Complexity", "abstract": "Conformal prediction has emerged as a widely used framework for constructing valid prediction sets in classification and regression tasks. In this work, we extend the split conformal prediction framework to hierarchical classification, where prediction sets are commonly restricted to internal nodes of a predefined hierarchy, and propose two computationally efficient inference algorithms. The first algorithm returns internal nodes as prediction sets, while the second one relaxes this restriction. Using the notion of representation complexity, the latter yields smaller set sizes at the cost of a more general and combinatorial inference problem. Empirical evaluations on several benchmark datasets demonstrate the effectiveness of the proposed algorithms in achieving nominal coverage.", "date": "2025-01-31", "categories": []} {"id": "2604.06468", "title": "Conformal Margin Risk Minimization: An Envelope Framework for Robust Learning under Label Noise", "abstract": "Most methods for learning with noisy labels require privileged knowledge such as noise transition matrices, clean subsets or pretrained feature extractors, resources typically unavailable when robustness is most needed. We propose Conformal Margin Risk Minimization (CMRM), a plug-and-play envelope framework that improves any classification loss under label noise by adding a single quantile-calibrated regularization term, with no privileged knowledge or training pipeline modification. CMRM measures the confidence margin between the observed label and competing labels, and thresholds it with a conformal quantile estimated per batch to focus training on high-margin samples while suppressing likely mislabeled ones. We derive a learning bound for CMRM under arbitrary label noise requiring only mild regularity of the margin distribution. Across five base methods and six benchmarks with synthetic and real-world noise, CMRM consistently improves accuracy (up to +3.39%), reduces conformal prediction set size (up to -20.44%) and does not hurt under 0% noise, showing that CMRM captures a method-agnostic uncertainty signal that existing mechanisms did not exploit.", "date": "2026-04-07", "categories": []} {"id": "2604.08885", "title": "Uncertainty-Aware Transformers: Conformal Prediction for Language Models", "abstract": "Transformers have had a profound impact on the field of artificial intelligence, especially on large language models and their variants. However, as was the case with neural networks, their black-box nature limits trust and deployment in high-stakes settings. For models to be genuinely useful and trustworthy in critical applications, they must provide more than just predictions: they must supply users with a clear understanding of the reasoning that underpins their decisions. This article presents an uncertainty quantification framework for transformer-based language models. This framework, called CONFIDE (CONformal prediction for FIne-tuned DEep language models), applies conformal prediction to the internal embeddings of encoder-only architectures, like BERT and RoBERTa, while enabling hyperparameter tuning. CONFIDE uses either [CLS] token embeddings or flattened hidden states to construct class-conditional nonconformity scores, enabling statistically valid prediction sets with instance-level explanations. Empirically, CONFIDE improves test accuracy by up to 4.09% on BERT-tiny and achieves greater correct efficiency (i.e., the expected size of the prediction set conditioned on it containing the true label) compared to prior methods, including NM2 and VanillaNN. We show that early and intermediate transformer layers often yield better-calibrated and more semantically meaningful representations for conformal prediction. In resource-constrained models and high-stakes tasks with ambiguous labels, CONFIDE offers robustness and interpretability where softmax-based uncertainty fails. We position CONFIDE as a framework for practical diagnostic and efficiency/robustness improvement over prior conformal baselines.", "date": "2026-04-10", "categories": []} {"id": "2604.07667", "title": "From Debate to Decision: Conformal Social Choice for Safe Multi-Agent Deliberation", "abstract": "Multi-agent debate improves LLM reasoning, yet agreement among agents is not evidence of correctness. When agents converge on a wrong answer through social reinforcement, consensus-based stopping commits that error to an automated action with no recourse. We introduce Conformal Social Choice, a post-hoc decision layer that converts debate outputs into calibrated act-versus-escalate decisions. Verbalized probability distributions from heterogeneous agents are aggregated via a linear opinion pool and calibrated with split conformal prediction, yielding prediction sets with a marginal coverage guarantee: the correct answer is included with probability ${\\geq}\\,1{-}\\alpha$, without assumptions on individual model calibration. A hierarchical action policy maps singleton sets to autonomous action and larger sets to human escalation. On eight MMLU-Pro domains with three agents (Claude Haiku, DeepSeek-R1, Qwen-3 32B), coverage stays within 1--2 points of the target. The key finding is not that debate becomes more accurate, but that the conformal layer makes its failures actionable: 81.9% of wrong-consensus cases are intercepted at $\\alpha{=}0.05$. Because the layer refuses to act on cases where debate is confidently wrong, the remaining conformal singletons reach 90.0--96.8% accuracy (up to 22.1pp above consensus stopping) -- a selection effect, not a reasoning improvement. This safety comes at the cost of automation, but the operating point is user-adjustable via $\\alpha$.", "date": "2026-04-09", "categories": []} {"id": "2604.01413", "title": "Adaptive Stopping for Multi-Turn LLM Reasoning", "abstract": "Large Language Models (LLMs) increasingly rely on multi-turn reasoning and interaction, such as adaptive retrieval-augmented generation (RAG) and ReAct-style agents, to answer difficult questions. These methods improve accuracy by iteratively retrieving information, reasoning, or acting, but introduce a key challenge: \\textbf{When should the model stop?} Existing approaches rely on heuristic stopping rules or fixed turn budgets and provide no formal guarantees that the final prediction still contains the correct answer. This limitation is particularly problematic in high-stakes domains such as finance and healthcare, where unnecessary turns increase cost and latency, while stopping too early risks incorrect decisions. Conformal prediction (CP) provides formal coverage guarantees, but existing LLM-CP methods only apply to a single model output and cannot handle multi-turn pipelines with adaptive stopping. To address this gap, we propose Multi-Turn Language Models with Conformal Prediction (MiCP), the first CP framework for multi-turn reasoning. MiCP allocates different error budgets across turns, enabling the model to stop early while maintaining an overall coverage guarantee. We demonstrate MiCP on adaptive RAG and ReAct, where it achieves the target coverage on both single-hop and multi-hop question answering benchmarks while reducing the number of turns, inference cost, and prediction set size. We further introduce a new metric that jointly evaluates coverage validity and answering efficiency.", "date": "2026-04-01", "categories": []} {"id": "2604.03772", "title": "Debiased Machine Learning for Conformal Prediction of Counterfactual Outcomes Under Runtime Confounding", "abstract": "Data-driven decision making frequently relies on predicting counterfactual outcomes. In practice, researchers commonly train counterfactual prediction models on a source dataset to inform decisions on a possibly separate target population. Conformal prediction has arisen as a popular method for producing assumption-lean prediction intervals for counterfactual outcomes that would arise under different treatment decisions in the target population of interest. However, existing methods require that every confounding factor of the treatment-outcome relationship used for training on the source data is additionally measured in the target population, risking miscoverage if important confounders are unmeasured in the target population. In this paper, we introduce a computationally efficient debiased machine learning framework that allows for valid prediction intervals when only a subset of confounders is measured in the target population, a common challenge referred to as runtime confounding. Grounded in semiparametric efficiency theory, we show the resulting prediction intervals achieve desired coverage rates with faster convergence compared to standard methods. Through numerous synthetic and semi-synthetic experiments, we demonstrate the utility of our proposed method.", "date": "2026-04-04", "categories": []} {"id": "2510.04318", "title": "Adaptive Coverage Policies in Conformal Prediction", "abstract": "Traditional conformal prediction methods construct prediction sets such that the true label falls within the set with a user-specified coverage level. However, poorly chosen coverage levels can result in uninformative predictions, either producing overly conservative sets when the coverage level is too high, or empty sets when it is too low. Moreover, the fixed coverage level cannot adapt to the specific characteristics of each individual example, limiting the flexibility and efficiency of these methods. In this work, we leverage recent advances in e-values and post-hoc conformal inference, which allow the use of data-dependent coverage levels while maintaining valid statistical guarantees. We propose to optimize an adaptive coverage policy by training a neural network using a leave-one-out procedure on the calibration set, allowing the coverage level and the resulting prediction set size to vary with the difficulty of each individual example. We support our approach with theoretical coverage guarantees and demonstrate its practical benefits through a series of experiments.", "date": "2025-10-05", "categories": []} {"id": "2510.25770", "title": "E-Scores for (In)Correctness Assessment of Generative Model Outputs", "abstract": "While generative models, especially large language models (LLMs), are ubiquitous in today's world, principled mechanisms to assess their (in)correctness are limited. Using the conformal prediction framework, previous works construct sets of LLM responses where the probability of including an incorrect response, or error, is capped at a user-defined tolerance level. However, since these methods are based on p-values, they are susceptible to p-hacking, i.e., choosing the tolerance level post-hoc can invalidate the guarantees. We therefore leverage e-values to complement generative model outputs with e-scores as measures of incorrectness. In addition to achieving the guarantees as before, e-scores further provide users with the flexibility of choosing data-dependent tolerance levels while upper bounding size distortion, a post-hoc notion of error. We experimentally demonstrate their efficacy in assessing LLM outputs under different forms of correctness: mathematical factuality and property constraints satisfaction.", "date": "2025-10-29", "categories": []} {"id": "2604.01170", "title": "Online Reasoning Calibration: Test-Time Training Enables Generalizable Conformal LLM Reasoning", "abstract": "While test-time scaling has enabled large language models to solve highly difficult tasks, state-of-the-art results come at exorbitant compute costs. These inefficiencies can be attributed to the miscalibration of post-trained language models, and the lack of calibration in popular sampling techniques. Here, we present Online Reasoning Calibration (ORCA), a framework for calibrating the sampling process that draws upon conformal prediction and test-time training. Specifically, we introduce a meta-learning procedure that updates the calibration module for each input. This allows us to provide valid confidence estimates under distributional shift, e.g. in thought patterns that occur across different stages of reasoning, or in prompt distributions between model development and deployment. ORCA not only provides theoretical guarantees on conformal risks, but also empirically shows higher efficiency and generalization across different reasoning tasks. At risk level $\\delta=0.1$, ORCA improves Qwen2.5-32B efficiency on in-distribution tasks with savings up to 47.5% with supervised labels and 40.7% with self-consistency labels. Under zero-shot out-of-domain settings, it improves MATH-500 savings from 24.8% of the static calibration baseline to 67.0% while maintaining a low empirical error rate, and the same trend holds across model families and downstream benchmarks. Our code is publicly available at https://github.com/wzekai99/ORCA.", "date": "2026-04-01", "categories": []} {"id": "2505.12578", "title": "Stacked conformal prediction", "abstract": "We consider a method for conformalizing a stacked ensemble of predictive models, showing that the potentially simple form of the meta-learner at the top of the stack enables a procedure with manageable computational cost that achieves approximate marginal validity without requiring the use of a separate calibration sample. Empirical results indicate that the method compares favorably to a standard inductive alternative.", "date": "2025-05-18", "categories": []} {"id": "2509.12573", "title": "No Need for Learning to Defer? A Training Free Deferral Framework to Multiple Experts through Conformal Prediction", "abstract": "AI systems often struggle to provide reliable predictions across all inputs, motivating hybrid human-AI decision-making. Existing Learning to Defer (L2D) approaches address this by training models to selectively defer to human experts. However, these approaches require extensive training data annotated by all experts and are sensitive to changes in expert composition, necessitating costly retraining. We propose a training-free, model- and expert-agnostic framework for expert deferral based on conformal prediction. Our method leverages prediction sets from a conformal predictor to quantify label-specific uncertainty and selects the most suitable expert using a segregativity criterion, which measures how well an expert discriminates among plausible labels. Experiments across three models on CIFAR10-H and HAM10000 demonstrate that our method can reduce the number of training labels per expert by up to 91.3% while maintaining predictive accuracy in low-data regimes. Being training-free, it also reduces training time by two orders of magnitude, offering a scalable, alternative to L2D for real-world human-AI collaboration.", "date": "2025-09-16", "categories": []} {"id": "2510.10324", "title": "On some practical challenges of conformal prediction", "abstract": "Conformal prediction is a model-free machine learning method for constructing prediction regions at a guaranteed coverage probability level. However, a data scientist often faces three challenges in practice: (i) the determination of a conformal prediction region is only approximate, jeopardizing the finite-sample validity of prediction, (ii) the computation required could be prohibitively expensive, and (iii) the shape of a conformal prediction region is hard to control. This article offers new insights into the relationship among the monotonicity of the non-conformity measure, the monotonicity of the plausibility function, and the exact determination of a conformal prediction region. Based on these new insights, we propose a quadratic-polynomial non-conformity measure that allows a data scientist to circumvent the three challenges simultaneously within the full conformal prediction framework.", "date": "2025-10-11", "categories": []} {"id": "2602.07530", "title": "Compact Conformal Subgraphs", "abstract": "Conformal prediction provides rigorous, distribution-free uncertainty guarantees, but often yields prohibitively large prediction sets in structured domains such as routing, planning, or sequential recommendation. We introduce \"graph-based conformal compression\", a framework for constructing compact subgraphs that preserve statistical validity while reducing structural complexity. We formulate compression as selecting a smallest subgraph capturing a prescribed fraction of the probability mass, and reduce to a weighted version of densest $k$-subgraphs in hypergraphs, in the regime where the subgraph has a large fraction of edges. We design efficient approximation algorithms that achieve constant factor coverage and size trade-offs. Crucially, we prove that our relaxation satisfies a monotonicity property, derived from a connection to parametric minimum cuts, which guarantees the nestedness required for valid conformal guarantees. Our results on the one hand bridge efficient conformal prediction with combinatorial graph compression via monotonicity, to provide rigorous guarantees on both statistical validity, and compression or size. On the other hand, they also highlight an algorithmic regime, distinct from classical densest-$k$-subgraph hardness settings, where the problem can be approximated efficiently. We finally validate our algorithmic approach via simulations for trip planning and navigation, and compare to natural baselines.", "date": "2026-02-07", "categories": []} {"id": "2603.27403", "title": "Conditional Factuality Controlled LLMs with Generalization Certificates via Conformal Sampling", "abstract": "Large language models (LLMs) need reliable test-time control of hallucinations. Existing conformal methods for LLMs typically provide only \\emph{marginal} guarantees and rely on a single global threshold, which can under-cover hard prompts, over-cover easy ones, and produce oversized prediction sets. We propose \\emph{Conditional Factuality Control} (CFC), a post-hoc conformal framework that returns \\emph{set-valued} outputs with \\emph{conditional} coverage guarantees. CFC defines a continuous, feature-conditional acceptance threshold through augmented quantile regression on a latent ``success'' score, and deploys it through a fixed-point threshold rule at inference time. Theoretically, we show that CFC satisfies a conditional coverage guarantee under exchangeability and analyze its \\emph{efficiency}, proving that, under mild assumptions on the score distributions, the conditional rule is strictly more sample-efficient than marginal conformal prediction at the same target coverage. We further derive a PAC-style variant, CFC-PAC, which shrinks the nominal risk level based on a stability bound, yielding a finite-sample certificate that the conditional miscoverage deviates from the target by at most $O(\\sqrt{\\log(1/\\delta)/N})$. Empirically, on synthetic data, real-world reasoning and QA benchmarks, and a Flickr8k VLM setting, CFC and CFC-PAC consistently attain near-target coverage across difficulty groups while using smaller prediction sets than CP and non-CP baselines.", "date": "2026-03-28", "categories": []} {"id": "2603.02460", "title": "Conformal Graph Prediction with Z-Gromov Wasserstein Distances", "abstract": "Supervised graph prediction addresses regression problems where the outputs are structured graphs. Although several approaches exist for graph-valued prediction, principled uncertainty quantification remains limited. We propose a conformal prediction framework for graph-valued outputs, providing distribution-free coverage guarantees in structured output spaces. Our method defines nonconformity via the Z-Gromov-Wasserstein distance, instantiated in practice through Fused Gromov-Wasserstein (FGW), enabling permutation invariant comparison between predicted and candidate graphs. To obtain adaptive prediction sets, we introduce Score Conformalized Quantile Regression (SCQR), an extension of Conformalized Quantile Regression (CQR) to handle complex output spaces such as graph-valued outputs. We evaluate the proposed approach on a synthetic task.", "date": "2026-03-02", "categories": []} {"id": "2603.26261", "title": "Contrastive Conformal Sets", "abstract": "Contrastive learning produces coherent semantic feature embeddings by encouraging positive samples to cluster closely while separating negative samples. However, existing contrastive learning methods lack principled guarantees on coverage within the semantic feature space. We extend conformal prediction to this setting by introducing minimum-volume covering sets equipped with learnable generalized multi-norm constraints. We propose a method that constructs conformal sets guaranteeing user-specified coverage of positive samples while maximizing negative sample exclusion. We establish theoretically that volume minimization serves as a proxy for negative exclusion, enabling our approach to operate effectively even when negative pairs are unavailable. The positive inclusion guarantee inherits the distribution-free coverage property of conformal prediction, while negative exclusion is maximized through learned set geometry optimized on a held-out training split. Experiments on simulated and real-world image datasets demonstrate improved inclusion-exclusion trade-offs compared to standard distance-based conformal baselines.", "date": "2026-03-27", "categories": []} {"id": "2603.26415", "title": "KMM-CP: Practical Conformal Prediction under Covariate Shift via Selective Kernel Mean Matching", "abstract": "Uncertainty quantification is essential for deploying machine learning models in high-stakes domains such as scientific discovery and healthcare. Conformal Prediction (CP) provides finite-sample coverage guarantees under exchangeability, an assumption often violated in practice due to distribution shift. Under covariate shift, restoring validity requires importance weighting, yet accurate density-ratio estimation becomes unstable when training and test distributions exhibit limited support overlap. We propose KMM-CP, a conformal prediction framework based on Kernel Mean Matching (KMM) for covariate-shift correction. We show that KMM directly controls the bias-variance components governing conformal coverage error by minimizing RKHS moment discrepancy under explicit weight constraints, and establish asymptotic coverage guarantees under mild conditions. We then introduce a selective extension that identifies regions of reliable support overlap and restricts conformal correction to this subset, further improving stability in low-overlap regimes. Experiments on molecular property prediction benchmarks with realistic distribution shifts show that KMM-CP reduces coverage gap by over 50% compared to existing approaches. The code is available at https://github.com/siddharthal/KMM-CP.", "date": "2026-03-27", "categories": []} {"id": "2603.24704", "title": "Conformal Selective Prediction with General Risk Control", "abstract": "In deploying artificial intelligence (AI) models, selective prediction offers the option to abstain from making a prediction when uncertain about model quality. To fulfill its promise, it is crucial to enforce strict and precise error control over cases where the model is trusted. We propose Selective Conformal Risk control with E-values (SCoRE), a new framework for deriving such decisions for any trained model and any user-defined, bounded and continuously-valued risk. SCoRE offers two types of guarantees on the risk among ``positive'' cases in which the system opts to trust the model. Built upon conformal inference and hypothesis testing ideas, SCoRE first constructs a class of (generalized) e-values, which are non-negative random variables whose product with the unknown risk has expectation no greater than one. Such a property is ensured by data exchangeability without requiring any modeling assumptions. Passing these e-values on to hypothesis testing procedures, we yield the binary trust decisions with finite-sample error control. SCoRE avoids the need of uniform concentration, and can be readily extended to settings with distribution shifts. We evaluate the proposed methods with simulations and demonstrate their efficacy through applications to error management in drug discovery, health risk prediction, and large language models.", "date": "2026-03-25", "categories": []} {"id": "2603.23919", "title": "Uncertainty-Aware Vision-based Risk Object Identification via Conformal Risk Tube Prediction", "abstract": "We study object importance-based vision risk object identification (Vision-ROI), a key capability for hazard detection in intelligent driving systems. Existing approaches make deterministic decisions and ignore uncertainty, which could lead to safety-critical failures. Specifically, in ambiguous scenarios, fixed decision thresholds may cause premature or delayed risk detection and temporally unstable predictions, especially in complex scenes with multiple interacting risks. Despite these challenges, current methods lack a principled framework to model risk uncertainty jointly across space and time. We propose Conformal Risk Tube Prediction, a unified formulation that captures spatiotemporal risk uncertainty, provides coverage guarantees for true risks, and produces calibrated risk scores with uncertainty estimates. To conduct a systematic evaluation, we present a new dataset and metrics probing diverse scenario configurations with multi-risk coupling effects, which are not supported by existing datasets. We systematically analyze factors affecting uncertainty estimation, including scenario variations, per-risk category behavior, and perception error propagation. Our method delivers substantial improvements over prior approaches, enhancing vision-ROI robustness and downstream performance, such as reducing nuisance braking alerts. For more qualitative results, please visit our project webpage: https://hcis-lab.github.io/CRTP/", "date": "2026-03-25", "categories": []} {"id": "2603.24475", "title": "Conformalized Transfer Learning for Li-ion Battery State of Health Forecasting under Manufacturing and Usage Variability", "abstract": "Accurate forecasting of state-of-health (SOH) is essential for ensuring safe and reliable operation of lithium-ion cells. However, existing models calibrated on laboratory tests at specific conditions often fail to generalize to new cells that differ due to small manufacturing variations or operate under different conditions. To address this challenge, an uncertainty-aware transfer learning framework is proposed, combining a Long Short-Term Memory (LSTM) model with domain adaptation via Maximum Mean Discrepancy (MMD) and uncertainty quantification through Conformal Prediction (CP). The LSTM model is trained on a virtual battery dataset designed to capture real-world variability in electrode manufacturing and operating conditions. MMD aligns latent feature distributions between simulated and target domains to mitigate domain shift, while CP provides calibrated, distribution-free prediction intervals. This framework improves both the generalization and trustworthiness of SOH forecasts across heterogeneous cells.", "date": "2026-03-25", "categories": []} {"id": "2603.22331", "title": "Conformal Risk Control for Safety-Critical Wildfire Evacuation Mapping: A Comparative Study of Tabular, Spatial, and Graph-Based Models", "abstract": "Every wildfire prediction model deployed today shares a dangerous property: none of these methods provides formal guarantees on how much fire spread is missed. Despite extensive work on wildfire spread prediction using deep learning, no prior study has applied distribution-free safety guarantees to this domain, leaving evacuation planners reliant on probability thresholds with no formal assurance. We address this gap by presenting, to our knowledge, the first application of conformal risk control (CRC) to wildfire spread prediction, providing finite-sample guarantees on false negative rate (FNR <= 0.05). We expose a stark failure: across three model families of increasing complexity (tabular: LightGBM, AUROC 0.854; convolutional: Tiny U-Net, AUROC 0.969; and graph-based: Hybrid ResGNN-UNet, AUROC 0.964), standard thresholds capture only 7-72% of true fire spread. CRC eliminates this failure uniformly. Our central finding is that model architecture determines evacuation efficiency, while CRC determines safety: both spatial models with CRC achieve approximately 95% fire coverage while flagging only approximately 15% of total pixels, making them 4.2x more efficient than LightGBM, while the graph model's additional complexity over a simple U-Net yields no meaningful efficiency gain. We propose a shift-aware three-way CRC framework that assigns SAFE/MONITOR/EVACUATE zones for operational triage, and characterize a fundamental limitation of prevalence-weighted bounds under extreme class imbalance (approximately 5% fire prevalence). All models, calibration code, and evaluation pipelines are released for reproducibility.", "date": "2026-03-20", "categories": []} {"id": "2603.23205", "title": "Between Resolution Collapse and Variance Inflation: Weighted Conformal Anomaly Detection in Low-Data Regimes", "abstract": "Standard conformal anomaly detection provides marginal finite-sample guarantees under the assumption of exchangeability . However, real-world data often exhibit distribution shifts, necessitating a weighted conformal approach to adapt to local non-stationarity. We show that this adaptation induces a critical trade-off between the minimum attainable p-value and its stability. As importance weights localize to relevant calibration instances, the effective sample size decreases. This can render standard conformal p-values overly conservative for effective error control, while the smoothing technique used to mitigate this issue introduces conditional variance, potentially masking anomalies. We propose a continuous inference relaxation that resolves this dilemma by decoupling local adaptation from tail resolution via continuous weighted kernel density estimation. While relaxing finite-sample exactness to asymptotic validity, our method eliminates Monte Carlo variability and recovers the statistical power lost to discretization. Empirical evaluations confirm that our approach not only restores detection capabilities where discrete baselines yield zero discoveries, but outperforms standard methods in statistical power while maintaining valid marginal error control in practice.", "date": "2026-03-24", "categories": []} {"id": "2503.04071", "title": "Tightening optimality gap with confidence through conformal prediction", "abstract": "Decision makers routinely use constrained optimization technology to plan and operate complex systems like global supply chains or power grids. In this context, practitioners must assess how close a computed solution is to optimality in order to make operational decisions, such as whether the current solution is sufficient or whether additional computation is warranted. A common practice is to evaluate solution quality using dual bounds returned by optimization solvers. While these dual bounds come with certified guarantees, they are often too loose to be practically informative. To this end, this paper introduces a novel conformal prediction framework for tightening loose primal and dual bounds. The proposed method addresses the heteroskedasticity commonly observed in these bounds via selective inference, and further exploits their inherent certified validity to produce tighter, more informative prediction intervals. Finally, numerical experiments on large-scale industrial problems suggest that the proposed approach can provide the same coverage level more efficiently than baseline methods.", "date": "2025-03-06", "categories": []} {"id": "2502.05709", "title": "Flow-based Conformal Prediction for Multi-dimensional Time Series", "abstract": "Time series prediction underpins a broad range of downstream tasks across many scientific domains. Recent advances and increasing adoption of black-box machine learning models for time series prediction highlight the critical need for uncertainty quantification. While conformal prediction has gained attention as a reliable uncertainty quantification method, conformal prediction for time series faces two key challenges: (1) \\textbf{leveraging correlations in observations and non-conformity scores to overcome the exchangeability assumption}, and (2) \\textbf{constructing prediction sets for multi-dimensional outcomes}. To address these challenges, we propose a novel conformal prediction method for time series using flow with classifier-free guidance. We provide coverage guarantees by establishing exact non-asymptotic marginal coverage and a finite-sample bound on conditional coverage for the proposed method. Evaluations on real-world time series datasets demonstrate that our method constructs significantly smaller prediction sets than existing conformal prediction methods, maintaining target coverage.", "date": "2025-02-08", "categories": []} {"id": "2503.19068", "title": "Minimum Volume Conformal Sets for Multivariate Regression", "abstract": "Conformal prediction provides a principled framework for constructing predictive sets with finite-sample validity. While much of the focus has been on univariate response variables, existing multivariate methods either impose rigid geometric assumptions or rely on flexible but computationally expensive approaches that do not explicitly optimize prediction set volume. We propose an optimization-driven framework based on a novel loss function that directly learns minimum-volume covering sets while ensuring valid coverage. This formulation naturally induces a new nonconformity score for conformal prediction, which adapts to the residual distribution and covariates. Our approach optimizes over prediction sets defined by arbitrary norm balls, including single and multi-norm formulations. Additionally, by jointly optimizing both the predictive model and predictive uncertainty, we obtain prediction sets that are tight, informative, and computationally efficient, as demonstrated in our experiments on real-world datasets.", "date": "2025-03-24", "categories": []} {"id": "2603.14198", "title": "Efficient Federated Conformal Prediction with Group-Conditional Guarantees", "abstract": "Deploying trustworthy AI systems requires principled uncertainty quantification. Conformal prediction (CP) is a widely used framework for constructing prediction sets with distribution-free coverage guarantees. In many practical settings, including healthcare, finance, and mobile sensing, the calibration data required for CP are distributed across multiple clients, each with its own local data distribution. In this federated setting, data can often be partitioned into, potentially overlapping, groups, which may reflect client-specific strata or cross-cutting attributes such as demographic or semantic categories. We propose group-conditional federated conformal prediction (GC-FCP), a novel protocol that provides group-conditional coverage guarantees. GC-FCP constructs mergeable, group-stratified coresets from local calibration scores, enabling clients to communicate compact weighted summaries that support efficient aggregation and calibration at the server. Experiments on synthetic and real-world datasets validate the performance of GC-FCP compared to centralized calibration baselines.", "date": "2026-03-15", "categories": []} {"id": "2603.14623", "title": "Proactive Routing to Interpretable Surrogates with Distribution-Free Safety Guarantees", "abstract": "Model routing determines whether to use an accurate black-box model or a simpler surrogate that approximates it at lower cost or greater interpretability. In deployment settings, practitioners often wish to restrict surrogate use to inputs where its degradation relative to a reference model is controlled. We study proactive (input-based) routing, in which a lightweight gate selects the model before either runs, enabling distribution-free control of the fraction of routed inputs whose degradation exceeds a tolerance {\\tau}. The gate is trained to distinguish safe from unsafe inputs, and a routing threshold is chosen via Clopper-Pearson conformal calibration on a held-out set, guaranteeing that the routed-set violation rate is at most {\\alpha} with probability 1-{\\delta}. We derive a feasibility condition linking safe routing to the base safe rate {\\pi} and risk budget {\\alpha}, along with sufficient AUC thresholds ensuring that feasible routing exists. Across 35 OpenML datasets and multiple black-box model families, gate-based conformal routing maintains controlled violation while achieving substantially higher coverage than regression conformal and naive baselines. We further show that probabilistic calibration primarily affects routing efficiency rather than distribution-free validity.", "date": "2026-03-15", "categories": []} {"id": "2510.08724", "title": "Counterfactually Fair Conformal Prediction", "abstract": "While counterfactual fairness of point predictors is well studied, its extension to prediction sets--central to fair decision-making under uncertainty--remains underexplored. On the other hand, conformal prediction (CP) provides efficient, distribution-free, finite-sample valid prediction sets, yet does not ensure counterfactual fairness. We close this gap by developing Counterfactually Fair Conformal Prediction (CF-CP) that produces counterfactually fair prediction sets. Through symmetrization of conformity scores across protected-attribute interventions, we prove that CF-CP results in counterfactually fair prediction sets while maintaining the marginal coverage property. Furthermore, we empirically demonstrate that on both synthetic and real datasets, across regression and classification tasks, CF-CP achieves the desired counterfactual fairness and meets the target coverage rate with minimal increase in prediction set size. CF-CP offers a simple, training-free route to counterfactually fair uncertainty quantification.", "date": "2025-10-09", "categories": []} {"id": "2603.11134", "title": "Conformal e-prediction in the presence of confounding", "abstract": "This note extends conformal e-prediction to cover the case where there is observed confounding between the random object $X$ and its label $Y$. We consider both the case where the observed data is IID and a case where some dependence between observations is permitted.", "date": "2026-03-11", "categories": []} {"id": "2602.18045", "title": "Conformal Tradeoffs: Operational Profiles Beyond Coverage", "abstract": "Conformal prediction gives exact finite-sample coverage guarantees under exchangeability, but deployed systems are judged by more than coverage alone. For a fixed calibrated rule reused over a finite operational window, stakeholders also care about deployment-facing quantities such as commitment frequency, deferral, and decisive error exposure. These are not determined by coverage: calibration choices with similar coverage can still induce materially different operational profiles. We study this characterization gap in a scoped setting: binary split conformal prediction under exchangeability with a fixed deployed rule. We introduce the Small-Sample Beta Correction (SSBC) which gives finite-sample coverage semantics for the deployed rule: it inverts the Beta/Beta--Binomial law governing calibration-conditional coverage to map a user request $(\\alpha^\\star,\\delta)$ to the least conservative calibration grid point with calibration-conditional PAC semantics for the realized deployed rule. Calibrate-and-Audit then fixes the rule by calibration and uses an independent audit split to estimate the induced region--class label table, a reusable summary from which deployment-facing Key Performance Indicators (KPIs) follow by projection. Under this design, fixed operational rates admit exact finite-sample Binomial inference, while Beta--Binomial envelopes serve as practical predictive summaries for future windows. The induced partition also exposes regime boundaries, Pareto-relevant tradeoffs, and inverse-pricing questions for fixed downstream conventions. Simulations validate the SSBC semantics and compare audit-based summaries with leave-one-out planning proxies; molecular toxicity data provide an audit-based empirical example, and a solubility case study illustrates scenario planning once coverage semantics are fixed.", "date": "2026-02-20", "categories": []} {"id": "2603.07371", "title": "ConfHit: Conformal Generative Design with Oracle Free Guarantees", "abstract": "The success of deep generative models in scientific discovery requires not only the ability to generate novel candidates but also reliable guarantees that these candidates indeed satisfy desired properties. Recent conformal-prediction methods offer a path to such guarantees, but its application to generative modeling in drug discovery is limited by budget constraints, lack of oracle access, and distribution shift. To this end, we introduce ConfHit, a distribution-free framework that provides validity guarantees under these conditions. ConfHit formalizes two central questions: (i) Certification: whether a generated batch can be guaranteed to contain at least one hit with a user-specified confidence level, and (ii) Design: whether the generation can be refined to a compact set without weakening this guarantee. ConfHit leverages weighted exchangeability between historical and generated samples to eliminate the need for an experimental oracle, constructs multiple-sample density-ratio weighted conformal p-value to quantify statistical confidence in hits, and proposes a nested testing procedure to certify and refine candidate sets of multiple generated samples while maintaining statistical guarantees. Across representative generative molecule design tasks and a broad range of methods, ConfHit consistently delivers valid coverage guarantees at multiple confidence levels while maintaining compact certified sets, establishing a principled and reliable framework for generative modeling.", "date": "2026-03-07", "categories": []} {"id": "2505.21147", "title": "Semi-Supervised Conformal Prediction With Unlabeled Nonconformity Score", "abstract": "Conformal prediction (CP) is a powerful framework for uncertainty quantification, generating prediction sets with coverage guarantees. Split conformal prediction relies on labeled data in the calibration procedure. However, the labeled data is often limited in real-world scenarios, leading to unstable coverage performance in different runs. To address this issue, we extend CP to the semi-supervised setting and propose SemiCP, a new paradigm that leverages both labeled and unlabeled data for calibration. To achieve this, we introduce an unlabeled nonconformity score, Nearest Neighbor Matching (NNM) score. Specifically, NNM estimates the nonconformity scores of unlabeled samples using their most similar pseudo-labeled counterparts during calibration, while maintaining the original scores for labeled data. Theoretically, we demonstrate that the average coverage gap (i.e., the absolute difference between the empirical marginal coverage and the target coverage) of SemiCP can decrease significantly at a rate $\\mathcal{O}(1/\\sqrt{N})$ and converge to an error term, where $N$ is the number of unlabeled data. Extensive experiments validate the effectiveness of SemiCP under limited labeled data, reducing the average coverage gap by up to 77% on common benchmarks with 4000 unlabeled examples, when there are only 20 labeled examples.", "date": "2025-05-27", "categories": []} {"id": "2603.06826", "title": "CREDO: Epistemic-Aware Conformalized Credal Envelopes for Regression", "abstract": "Conformal prediction delivers prediction intervals with distribution-free coverage, but its intervals can look overconfident in regions where the model is extrapolating, because standard conformal scores do not explicitly represent epistemic uncertainty. Credal methods, by contrast, make epistemic effects visible by working with sets of plausible predictive distributions, but they are typically model-based and lack calibration guarantees. We introduce CREDO, a simple \"credal-then-conformalize\" recipe that combines both strengths. CREDO first builds an interpretable credal envelope that widens when local evidence is weak, then applies split conformal calibration on top of this envelope to guarantee marginal coverage without further assumptions. This separation of roles yields prediction intervals that are interpretable: their width can be decomposed into aleatoric noise, epistemic inflation, and a distribution-free calibration slack. We provide a fast implementation based on trimming extreme posterior predictive endpoints, prove validity, and show on benchmark regressions that CREDO maintains target coverage while improving sparsity adaptivity at competitive efficiency.", "date": "2026-03-06", "categories": []} {"id": "2510.07093", "title": "Non-Asymptotic Analysis of Efficiency in Conformalized Regression", "abstract": "Conformal prediction provides prediction sets with coverage guarantees. The informativeness of conformal prediction depends on its efficiency, typically quantified by the expected size of the prediction set. Prior work on the efficiency of conformalized regression commonly treats the miscoverage level $\\alpha$ as a fixed constant. In this work, we establish non-asymptotic bounds on the deviation of the prediction set length from the oracle interval length for conformalized quantile and median regression trained via SGD, under mild assumptions on the data distribution. Our bounds of order $\\mathcal{O}(1/\\sqrt{n} + 1/(\\alpha^2 n) + 1/\\sqrt{m} + \\exp(-\\alpha^2 m))$ capture the joint dependence of efficiency on the proper training set size $n$, the calibration set size $m$, and the miscoverage level $\\alpha$. The results identify phase transitions in convergence rates across different regimes of $\\alpha$, offering guidance for allocating data to control excess prediction set length. Empirical results are consistent with our theoretical findings.", "date": "2025-10-08", "categories": []} {"id": "2512.06945", "title": "Symmetric Aggregation of Conformity Scores for Efficient Uncertainty Sets", "abstract": "Access to multiple predictive models trained for the same task, whether in regression or classification, is increasingly common in many applications. Aggregating their predictive uncertainties to produce reliable and efficient uncertainty quantification is therefore a critical but still underexplored challenge, especially within the framework of conformal prediction (CP). While CP methods can generate individual prediction sets from each model, combining them into a single, more informative set remains a challenging problem. To address this, we propose SACP (Symmetric Aggregated Conformal Prediction), a novel method that aggregates nonconformity scores from multiple predictors. SACP transforms these scores into e-values and combines them using any symmetric aggregation function. This flexible design enables a robust, data-driven framework for selecting aggregation strategies that yield sharper prediction sets. We also provide theoretical insights that help justify the validity and performance of the SACP approach. Extensive experiments on diverse datasets show that SACP consistently improves efficiency and often outperforms state-of-the-art model aggregation baselines.", "date": "2025-12-07", "categories": []} {"id": "2602.16537", "title": "Optimal training-conditional regret for online conformal prediction", "abstract": "We study online conformal prediction for non-stationary data streams subject to unknown distribution drift. While most prior work studied this problem under adversarial settings and/or assessed performance in terms of gaps of time-averaged marginal coverage, we instead evaluate performance through training-conditional cumulative regret. We specifically focus on independently generated data with two types of distribution shift: abrupt change points and smooth drift.\n When non-conformity score functions are pretrained on an independent dataset, we propose a split-conformal style algorithm that leverages drift detection to adaptively update calibration sets, which provably achieves minimax-optimal regret. When non-conformity scores are instead trained online, we develop a full-conformal style algorithm that again incorporates drift detection to handle non-stationarity; this approach relies on stability - rather than permutation symmetry - of the model-fitting algorithm, which is often better suited to online learning under evolving environments. We establish non-asymptotic regret guarantees for our online full conformal algorithm, which match the minimax lower bound under appropriate restrictions on the prediction sets. Numerical experiments corroborate our theoretical findings.", "date": "2026-02-18", "categories": []} {"id": "2508.11025", "title": "Zono-Conformal Prediction: Zonotope-Based Uncertainty Quantification for Regression and Classification Tasks", "abstract": "Conformal prediction is a popular uncertainty quantification method that augments a base predictor to return sets of predictions with statistically valid coverage guarantees. However, current methods are often computationally expensive and data-intensive, as they require constructing an uncertainty model before calibration. Moreover, existing approaches typically represent the prediction sets with intervals, which limits their ability to capture dependencies in multi-dimensional outputs. We address these limitations by introducing zono-conformal prediction, a novel approach inspired by interval predictor models and reachset-conformant identification that constructs prediction zonotopes with assured coverage. By placing zonotopic uncertainty sets directly into the model of the base predictor, zono-conformal predictors can be identified via a single, data-efficient linear program. While we can apply zono-conformal prediction to arbitrary nonlinear base predictors, we focus on feed-forward neural networks in this work. Aside from regression tasks, we also construct optimal zono-conformal predictors in classification settings where the output of an uncertain predictor is a set of possible classes. We provide probabilistic coverage guarantees and present methods for detecting outliers in the identification data. In extensive numerical experiments, we show that zono-conformal predictors are less conservative than interval predictor models and standard conformal prediction methods, while achieving a similar coverage over the test data.", "date": "2025-08-14", "categories": []} {"id": "2509.22240", "title": "COMPASS: Robust Feature Conformal Prediction for Medical Segmentation Metrics", "abstract": "In clinical applications, the utility of segmentation models is often based on the accuracy of derived downstream metrics such as organ size, rather than by the pixel-level accuracy of the segmentation masks themselves. Thus, uncertainty quantification for such metrics is crucial for decision-making. Conformal prediction (CP) is a popular framework to derive such principled uncertainty guarantees, but applying CP naively to the final scalar metric is inefficient because it treats the complex, non-linear segmentation-to-metric pipeline as a black box. We introduce COMPASS, a practical framework that generates efficient, metric-based CP intervals for image segmentation models by leveraging the inductive biases of their underlying deep neural networks. COMPASS performs calibration directly in the model's representation space by perturbing intermediate features along low-dimensional subspaces maximally sensitive to the target metric. We prove that COMPASS achieves valid marginal coverage under the assumption of exchangeability. Empirically, we demonstrate that COMPASS produces significantly tighter intervals than traditional CP baselines on four medical image segmentation tasks for area estimation of skin lesions and anatomical structures. Furthermore, we show that leveraging learned internal features to estimate importance weights allows COMPASS to also recover target coverage under covariate shifts. COMPASS paves the way for practical, metric-based uncertainty quantification for medical image segmentation.", "date": "2025-09-26", "categories": []} {"id": "2510.05060", "title": "ResCP: Reservoir Conformal Prediction for Time Series Forecasting", "abstract": "Conformal prediction offers a powerful framework for building distribution-free prediction intervals for exchangeable data. Existing methods that extend conformal prediction to sequential data rely on fitting a relatively complex model to capture temporal dependencies. However, these methods can fail if the sample size is small and often require expensive retraining when the underlying data distribution changes. To overcome these limitations, we propose Reservoir Conformal Prediction (ResCP), a novel training-free conformal prediction method for time series. Our approach leverages the efficiency and representation learning capabilities of reservoir computing to dynamically reweight conformity scores. In particular, we compute similarity scores among reservoir states and use them to adaptively reweight the observed residuals at each step. With this approach, ResCP enables us to account for local temporal dynamics when modeling the error distribution without compromising computational scalability. We prove that, under reasonable assumptions, ResCP achieves asymptotic conditional coverage, and we empirically demonstrate its effectiveness across diverse forecasting tasks.", "date": "2025-10-06", "categories": []} {"id": "2603.00324", "title": "Proof-of-Perception: Certified Tool-Using Multimodal Reasoning with Compositional Conformal Guarantees", "abstract": "We present Proof-of-Perception (PoP), a tool-using framework that casts multimodal reasoning as an executable graph with explicit reliability guarantees. Each perception or logic node outputs a conformal set, yielding calibrated, stepwise uncertainty; a lightweight controller uses these certificates to allocate compute under a budget, expanding with extra tool calls only when needed and stopping early otherwise. This grounds answers in verifiable evidence, reduces error compounding and hallucinations, and enables principled accuracy-compute trade-offs. Across document, chart, and multi-image QA benchmarks, PoP improves performance and reliability over strong chain-of-thought, ReAct-style, and program-of-thought baselines while using computation more efficiently.", "date": "2026-02-27", "categories": []} {"id": "2603.01719", "title": "Co-optimization for Adaptive Conformal Prediction", "abstract": "Conformal prediction (CP) provides finite-sample, distribution-free marginal coverage, but standard conformal regression intervals can be inefficient under heteroscedasticity and skewness. In particular, popular constructions such as conformalized quantile regression (CQR) often inherit a fixed notion of center and enforce equal-tailed errors, which can displace the interval away from high-density regions and produce unnecessarily wide sets. We propose Co-optimization for Adaptive Conformal Prediction (CoCP), a framework that learns prediction intervals by jointly optimizing a center $m(x)$ and a radius $h(x)$.CoCP alternates between (i) learning $h(x)$ via quantile regression on the folded absolute residual around the current center, and (ii) refining $m(x)$ with a differentiable soft-coverage objective whose gradients concentrate near the current boundaries, effectively correcting mis-centering without estimating the full conditional density. Finite-sample marginal validity is guaranteed by split-conformal calibration with a normalized nonconformity score. Theory characterizes the population fixed point of the soft objective and shows that, under standard regularity conditions, CoCP asymptotically approaches the length-minimizing conditional interval at the target coverage level as the estimation error and smoothing vanish. Experiments on synthetic and real benchmarks demonstrate that CoCP yields consistently shorter intervals and achieves state-of-the-art conditional-coverage diagnostics.", "date": "2026-03-02", "categories": []} {"id": "2507.06867", "title": "Conformal Prediction for Long-Tailed Classification", "abstract": "Many real-world classification problems, such as plant identification, have extremely long-tailed class distributions. In order for prediction sets to be useful in such settings, they should (i) provide good class-conditional coverage, ensuring that rare classes are not systematically omitted from the prediction sets, and (ii) be a reasonable size, allowing users to easily verify candidate labels. Unfortunately, existing conformal prediction methods, when applied to the long-tailed setting, force practitioners to make a binary choice between small sets with poor class-conditional coverage or sets that have very good class-conditional coverage but are extremely large. We propose methods with marginal coverage guarantees that smoothly trade off set size and class-conditional coverage. First, we introduce a new conformal score function called prevalence-adjusted softmax that optimizes for macro-coverage, defined as the average class-conditional coverage across classes. Second, we propose a new procedure that interpolates between marginal and class-conditional conformal prediction by linearly interpolating their conformal score thresholds. We demonstrate our methods on Pl@ntNet-300K and iNaturalist-2018, two long-tailed image datasets with 1,081 and 8,142 classes, respectively.", "date": "2025-07-09", "categories": []} {"id": "2505.04733", "title": "Conformal Prediction with Corrupted Labels: Uncertain Imputation and Robust Re-weighting", "abstract": "We introduce a framework for robust uncertainty quantification in situations where labeled training data are corrupted, through noisy or missing labels. We build on conformal prediction, a statistical tool for generating prediction sets that cover the test label with a pre-specified probability. The validity of conformal prediction, however, holds under the i.i.d assumption, which does not hold in our setting due to the corruptions in the data. To account for this distribution shift, the privileged conformal prediction (PCP) method proposed leveraging privileged information (PI) -- additional features available only during training -- to re-weight the data distribution, yielding valid prediction sets under the assumption that the weights are accurate. In this work, we analyze the robustness of PCP to inaccuracies in the weights. Our analysis indicates that PCP can still yield valid uncertainty estimates even when the weights are poorly estimated. Furthermore, we introduce uncertain imputation (UI), a new conformal method that does not rely on weight estimation. Instead, we impute corrupted labels in a way that preserves their uncertainty. Our approach is supported by theoretical guarantees and validated empirically on both synthetic and real benchmarks. Finally, we show that these techniques can be integrated into a triply robust framework, ensuring statistically valid predictions as long as at least one underlying method is valid.", "date": "2025-05-07", "categories": []} {"id": "2602.22432", "title": "LoBoost: Fast Model-Native Local Conformal Prediction for Gradient-Boosted Trees", "abstract": "Gradient-boosted decision trees are among the strongest off-the-shelf predictors for tabular regression, but point predictions alone do not quantify uncertainty. Conformal prediction provides distribution-free marginal coverage, yet split conformal uses a single global residual quantile and can be poorly adaptive under heteroscedasticity. Methods that improve adaptivity typically fit auxiliary nuisance models or introduce additional data splits/partitions to learn the conformal score, increasing cost and reducing data efficiency. We propose LoBoost, a model-native local conformal method that reuses the fitted ensemble's leaf structure to define multiscale calibration groups. Each input is encoded by its sequence of visited leaves; at resolution level k, we group points by matching prefixes of leaf indices across the first k trees and calibrate residual quantiles within each group. LoBoost requires no retraining, auxiliary models, or extra splitting beyond the standard train/calibration split. Experiments show competitive interval quality, improved test MSE on most datasets, and large calibration speedups.", "date": "2026-02-25", "categories": []} {"id": "2602.21368", "title": "Black-Box Reliability Certification for AI Agents via Self-Consistency Sampling and Conformal Calibration", "abstract": "Given a black-box AI system and a task, at what confidence level can a practitioner trust the system's output? We answer with a reliability level -- a single number per system-task pair, derived from self-consistency sampling and conformal calibration, that serves as a black-box deployment gate with exact, finite-sample, distribution-free guarantees. Self-consistency sampling reduces uncertainty exponentially; conformal calibration guarantees correctness within 1/(n+1) of the target level, regardless of the system's errors -- made transparently visible through larger answer sets for harder questions. Weaker models earn lower reliability levels (not accuracy -- see Definition 2.4): GPT-4.1 earns 94.6% on GSM8K and 96.8% on TruthfulQA, while GPT-4.1-nano earns 89.8% on GSM8K and 66.5% on MMLU. We validate across five benchmarks, five models from three families, and both synthetic and real data. Conditional coverage on solvable items exceeds 0.93 across all configurations; sequential stopping reduces API costs by around 50%.", "date": "2026-02-24", "categories": []} {"id": "2504.13961", "title": "CONTINA: Confidence Interval for Traffic Demand Prediction with Coverage Guarantee", "abstract": "Accurate short-term traffic demand prediction is critical for the operation of traffic systems. Besides point estimation, the confidence interval of the prediction is also of great importance. Many models for traffic operations, such as shared bike rebalancing and taxi dispatching, take into account the uncertainty of future demand and require confidence intervals as the input. However, existing methods for confidence interval modeling rely on strict assumptions, such as unchanging traffic patterns and correct model specifications, to guarantee enough coverage. Therefore, the confidence intervals provided could be invalid, especially in a changing traffic environment. To fill this gap, we propose an efficient method, CONTINA (Conformal Traffic Intervals with Adaptation) to provide interval predictions that can adapt to external changes. By collecting the errors of interval during deployment, the method can adjust the interval in the next step by widening it if the errors are too large or shortening it otherwise. Furthermore, we theoretically prove that the coverage of the confidence intervals provided by our method converges to the target coverage level. Experiments across four real-world datasets and prediction models demonstrate that the proposed method can provide valid confidence intervals with shorter lengths. Our method can help traffic management personnel develop a more reasonable and robust operation plan in practice. And we release the code, model and dataset in \\href{ https://github.com/xiannanhuang/CONTINA/}{ Github}.", "date": "2025-04-17", "categories": []} {"id": "2512.07770", "title": "Distribution-informed Online Conformal Prediction", "abstract": "Conformal prediction provides a pivotal and flexible technique for uncertainty quantification by constructing prediction sets with a predefined coverage rate. Many online conformal prediction methods have been developed to address data distribution shifts in fully adversarial environments, resulting in overly conservative prediction sets. We propose Conformal Optimistic Prediction (COP), an online conformal prediction algorithm incorporating underlying data pattern into the update rule. Through estimated cumulative distribution function of non-conformity scores, COP produces tighter prediction sets when predictable pattern exists, while retaining valid coverage guarantees even when estimates are inaccurate. We establish a joint bound on coverage and regret, which further confirms the validity of our approach. We also prove that COP achieves distribution-free, finite-sample coverage under arbitrary learning rates and can converge when scores are $i.i.d.$. The experimental results also show that COP can achieve valid coverage and construct shorter prediction intervals than other baselines.", "date": "2025-12-08", "categories": []} {"id": "2602.20652", "title": "DANCE: Doubly Adaptive Neighborhood Conformal Estimation", "abstract": "The recent developments of complex deep learning models have led to unprecedented ability to accurately predict across multiple data representation types. Conformal prediction for uncertainty quantification of these models has risen in popularity, providing adaptive, statistically-valid prediction sets. For classification tasks, conformal methods have typically focused on utilizing logit scores. For pre-trained models, however, this can result in inefficient, overly conservative set sizes when not calibrated towards the target task. We propose DANCE, a doubly locally adaptive nearest-neighbor based conformal algorithm combining two novel nonconformity scores directly using the data's embedded representation. DANCE first fits a task-adaptive kernel regression model from the embedding layer before using the learned kernel space to produce the final prediction sets for uncertainty quantification. We test against state-of-the-art local, task-adapted and zero-shot conformal baselines, demonstrating DANCE's superior blend of set size efficiency and robustness across various datasets.", "date": "2026-02-24", "categories": []} {"id": "2602.19498", "title": "Softmax is not Enough (for Adaptive Conformal Classification)", "abstract": "The merit of Conformal Prediction (CP), as a distribution-free framework for uncertainty quantification, depends on generating prediction sets that are efficient, reflected in small average set sizes, while adaptive, meaning they signal uncertainty by varying in size according to input difficulty. A central limitation for deep conformal classifiers is that the nonconformity scores are derived from softmax outputs, which can be unreliable indicators of how certain the model truly is about a given input, sometimes leading to overconfident misclassifications or undue hesitation. In this work, we argue that this unreliability can be inherited by the prediction sets generated by CP, limiting their capacity for adaptiveness. We propose a new approach that leverages information from the pre-softmax logit space, using the Helmholtz Free Energy as a measure of model uncertainty and sample difficulty. By reweighting nonconformity scores with a monotonic transformation of the energy score of each sample, we improve their sensitivity to input difficulty. Our experiments with four state-of-the-art score functions on multiple datasets and deep architectures show that this energy-based enhancement improves the adaptiveness of the prediction sets, leading to a notable increase in both efficiency and adaptiveness compared to baseline nonconformity scores, without introducing any post-hoc complexity.", "date": "2026-02-23", "categories": []} {"id": "2602.20151", "title": "Conformal Risk Control for Non-Monotonic Losses", "abstract": "Conformal risk control is an extension of conformal prediction for controlling risk functions beyond miscoverage. The original algorithm controls the expected value of a loss that is monotonic in a one-dimensional parameter. Here, we present risk control guarantees for generic algorithms applied to possibly non-monotonic losses with multidimensional parameters. The guarantees depend on the stability of the algorithm -- unstable algorithms have looser guarantees. We give applications of this technique to selective image classification, FDR and IOU control of tumor segmentations, and multigroup debiasing of recidivism predictions across overlapping race and sex groups using empirical risk minimization.", "date": "2026-02-23", "categories": []} {"id": "2602.18396", "title": "PRISM-FCP: Byzantine-Resilient Federated Conformal Prediction via Partial Sharing", "abstract": "We propose PRISM-FCP (Partial shaRing and robust calIbration with Statistical Margins for Federated Conformal Prediction), a Byzantine-resilient federated conformal prediction framework that utilizes partial model sharing to improve robustness against Byzantine attacks during both model training and conformal calibration. Existing approaches address adversarial behavior only in the calibration stage, leaving the learned model susceptible to poisoned updates. In contrast, PRISM-FCP mitigates attacks end-to-end. During training, clients partially share updates by transmitting only $M$ of $D$ parameters per round. This attenuates the expected energy of an adversary's perturbation in the aggregated update by a factor of $M/D$, yielding lower mean-square error (MSE) and tighter prediction intervals. During calibration, clients convert nonconformity scores into characterization vectors, compute distance-based maliciousness scores, and downweight or filter suspected Byzantine contributions before estimating the conformal quantile. Extensive experiments on both synthetic data and the UCI Superconductivity dataset demonstrate that PRISM-FCP maintains nominal coverage guarantees under Byzantine attacks while avoiding the interval inflation observed in standard FCP with reduced communication, providing a robust and communication-efficient approach to federated uncertainty quantification.", "date": "2026-02-20", "categories": []} {"id": "2602.17535", "title": "LATA: Laplacian-Assisted Transductive Adaptation for Conformal Uncertainty in Medical VLMs", "abstract": "Medical vision-language models (VLMs) are strong zero-shot recognizers for medical imaging, but their reliability under domain shift hinges on calibrated uncertainty with guarantees. Split conformal prediction (SCP) offers finite-sample coverage, yet prediction sets often become large (low efficiency) and class-wise coverage unbalanced-high class-conditioned coverage gap (CCV), especially in few-shot, imbalanced regimes; moreover, naively adapting to calibration labels breaks exchangeability and voids guarantees. We propose \\texttt{\\textbf{LATA}} (Laplacian-Assisted Transductive Adaptation), a \\textit{training- and label-free} refinement that operates on the joint calibration and test pool by smoothing zero-shot probabilities over an image-image k-NN graph using a small number of CCCP mean-field updates, preserving SCP validity via a deterministic transform. We further introduce a \\textit{failure-aware} conformal score that plugs into the vision-language uncertainty (ViLU) framework, providing instance-level difficulty and label plausibility to improve prediction set efficiency and class-wise balance at fixed coverage. \\texttt{\\textbf{LATA}} is black-box (no VLM updates), compute-light (windowed transduction, no backprop), and includes an optional prior knob that can run strictly label-free or, if desired, in a label-informed variant using calibration marginals once. Across \\textbf{three} medical VLMs and \\textbf{nine} downstream tasks, \\texttt{\\textbf{LATA}} consistently reduces set size and CCV while matching or tightening target coverage, outperforming prior transductive baselines and narrowing the gap to label-using methods, while using far less compute. Comprehensive ablations and qualitative analyses show that \\texttt{\\textbf{LATA}} sharpens zero-shot predictions without compromising exchangeability.", "date": "2026-02-19", "categories": []} {"id": "2412.06082", "title": "Are foundation models for computer vision good conformal predictors?", "abstract": "Recent advances in self-supervision and contrastive learning have brought the performance of foundation models to unprecedented levels in a variety of tasks. Fueled by this progress, these models are becoming the prevailing approach for a wide array of real-world vision problems, including risk-sensitive and high-stakes applications. However, ensuring safe deployment in these scenarios requires a more comprehensive understanding of their uncertainty modeling capabilities, which has received little attention. In this work, we delve into the behaviour of vision and vision-language foundation models under Conformal Prediction (CP), a statistical framework that provides theoretical guarantees of marginal coverage of the true class. Across extensive experiments including popular vision classification benchmarks, well-known foundation vision models, and three CP methods, our findings reveal that foundation models are well-suited for conformalization procedures, particularly those integrating Vision Transformers. We also show that calibrating the confidence predictions of these models, a popular strategy to improve their uncertainty quantification, actually leads to efficiency degradation of the conformal set on adaptive CP methods. Furthermore, few-shot adaptation of Vision-Language Models (VLMs) to downstream tasks, whose popularity is surging, enhances conformal scores compared to zero-shot predictions. Last, our empirical study exposes APS as particularly promising in the context of vision foundation models, as it does not violate the marginal coverage guarantees across multiple challenging, yet realistic scenarios.", "date": "2024-12-08", "categories": []} {"id": "2506.13593", "title": "Calibrated Predictive Lower Bounds on Time-to-Unsafe-Sampling in LLMs", "abstract": "We introduce time-to-unsafe-sampling, a novel safety measure for generative models, defined as the number of generations required by a large language model (LLM) to trigger an unsafe (e.g., toxic) response. While providing a new dimension for prompt-adaptive safety evaluation, quantifying time-to-unsafe-sampling is challenging: unsafe outputs are often rare in well-aligned models and thus may not be observed under any feasible sampling budget. To address this challenge, we frame this estimation problem as one of survival analysis. We build on recent developments in conformal prediction and propose a novel calibration technique to construct a lower predictive bound (LPB) on the time-to-unsafe-sampling of a given prompt with rigorous coverage guarantees. Our key technical innovation is an optimized sampling-budget allocation scheme that improves sample efficiency while maintaining distribution-free guarantees. Experiments on both synthetic and real data support our theoretical results and demonstrate the practical utility of our method for safety risk assessment in generative AI models.", "date": "2025-06-16", "categories": []} {"id": "2510.10193", "title": "SAFER: Risk-Constrained Sample-then-Filter in Large Language Models", "abstract": "As large language models (LLMs) are increasingly deployed in risk-sensitive applications such as real-world open-ended question answering (QA), ensuring the trustworthiness of their outputs has become critical. Existing selective conformal prediction (SCP) methods provide statistical guarantees by constructing prediction sets with a constrained miscoverage rate for correct answers. However, prior works unrealistically assume that admissible answers for all instances can be obtained via finite sampling, even for open-ended QA scenarios that lack a fixed and finite solution space. To address this, we introduce a two-stage risk control framework comprising abstention-aware sampling and conformalized filtering (SAFER). Firstly, on a held-out calibration set, SAFER calibrates a sampling budget within the maximum sampling cap, using the Clopper-Pearson exact method at a user-desired risk level (i.e., the maximum allowable miscoverage rate of the sampling sets). If the risk level cannot be satisfied within the cap, we abstain; otherwise, the calibrated sampling budget becomes the minimum requirements at test time. Then, we employ calibration instances where correct answers are attainable under the calibrated budget and apply the conformal risk control method to determine a statistically valid uncertainty threshold, which filters unreliable distractors from the candidate set for each test data point. In this stage, SAFER introduces an additional risk level to guide the calculation of the threshold, thereby controlling the risk of correct answers being excluded. Furthermore, we show that SAFER is compatible with various task-specific admission criteria and calibration-test split ratios, highlighting its robustness and high data efficiency.", "date": "2025-10-11", "categories": []} {"id": "2602.13660", "title": "Optimized Certainty Equivalent Risk-Controlling Prediction Sets", "abstract": "In safety-critical applications such as medical image segmentation, prediction systems must provide reliability guarantees that extend beyond conventional expected loss control. While risk-controlling prediction sets (RCPS) offer probabilistic guarantees on the expected risk, they fail to capture tail behavior and worst-case scenarios that are crucial in high-stakes settings. This paper introduces optimized certainty equivalent RCPS (OCE-RCPS), a novel framework that provides high-probability guarantees on general optimized certainty equivalent (OCE) risk measures, including conditional value-at-risk (CVaR) and entropic risk. OCE-RCPS leverages upper confidence bounds to identify prediction set parameters that satisfy user-specified risk tolerance levels with provable reliability. We establish theoretical guarantees showing that OCE-RCPS satisfies the desired probabilistic constraint for loss functions such as miscoverage and false negative rate. Experiments on image segmentation demonstrate that OCE-RCPS consistently meets target satisfaction rates across various risk measures and reliability configurations, while OCE-CRC fails to provide probabilistic guarantees.", "date": "2026-02-14", "categories": []} {"id": "2602.12693", "title": "Leverage-Weighted Conformal Prediction", "abstract": "Split conformal prediction provides distribution-free prediction intervals with finite-sample marginal coverage, but produces constant-width intervals that overcover in low-variance regions and undercover in high-variance regions. Existing adaptive methods require training auxiliary models. We propose Leverage-Weighted Conformal Prediction (LWCP), which weights nonconformity scores by a function of the statistical leverage -- the diagonal of the hat matrix -- deriving adaptivity from the geometry of the design matrix rather than from auxiliary model fitting. We prove that LWCP preserves finite-sample marginal validity for any weight function; achieves asymptotically optimal conditional coverage at essentially no width cost when heteroscedasticity factors through leverage; and recovers the form and width of classical prediction intervals under Gaussian assumptions while retaining distribution-free guarantees. We further establish that randomized leverage approximations preserve coverage exactly with controlled width perturbation, and that vanilla CP suffers a persistent, sample-size-independent conditional coverage gap that LWCP eliminates. The method requires no hyperparameters beyond the choice of weight function and adds negligible computational overhead to vanilla CP. Experiments on synthetic and real data confirm the theoretical predictions, demonstrating substantial reductions in conditional coverage disparity across settings.", "date": "2026-02-13", "categories": []} {"id": "2505.13732", "title": "Backward Conformal Prediction", "abstract": "We introduce $\\textit{Backward Conformal Prediction}$, a method that guarantees conformal coverage while providing flexible control over the size of prediction sets. Unlike standard conformal prediction, which fixes the coverage level and allows the conformal set size to vary, our approach defines a rule that constrains how prediction set sizes behave based on the observed data, and adapts the coverage level accordingly. Our method builds on two key foundations: (i) recent results by Gauthier et al. [2025] on post-hoc validity using e-values, which ensure marginal coverage of the form $\\mathbb{P}(Y_{\\rm test} \\in \\hat C_n^{\\tilde{\\alpha}}(X_{\\rm test})) \\ge 1 - \\mathbb{E}[\\tilde{\\alpha}]$ up to a first-order Taylor approximation for any data-dependent miscoverage $\\tilde{\\alpha}$, and (ii) a novel leave-one-out estimator $\\hat{\\alpha}^{\\rm LOO}$ of the marginal miscoverage $\\mathbb{E}[\\tilde{\\alpha}]$ based on the calibration set, ensuring that the theoretical guarantees remain computable in practice. This approach is particularly useful in applications where large prediction sets are impractical such as medical diagnosis. We provide theoretical results and empirical evidence supporting the validity of our method, demonstrating that it maintains computable coverage guarantees while ensuring interpretable, well-controlled prediction set sizes.", "date": "2025-05-19", "categories": []} {"id": "2508.03245", "title": "Conformal Unlearning: A New Paradigm for Unlearning in Conformal Predictors", "abstract": "Conformal unlearning aims to ensure that a trained conformal predictor miscovers data points with specific shared characteristics, such as those from a particular label class, associated with a specific user, or belonging to a defined cluster, while maintaining valid coverage on the remaining data. Existing machine unlearning methods, which typically approximate a model retrained from scratch after removing the data to be forgotten, face significant challenges when applied to conformal unlearning. These methods often lack rigorous, uncertainty-aware statistical measures to evaluate unlearning effectiveness and exhibit a mismatch between their degraded performance on forgotten data and the frequency with which that data are still correctly covered by conformal predictors-a phenomenon we term ''fake conformal unlearning''. To address these limitations, we propose a new paradigm for conformal machine unlearning that provides finite-sample, uncertainty-aware guarantees on unlearning performance without relying on a retrained model as a reference. We formalize conformal unlearning to require high coverage on retained data and high miscoverage on forgotten data, introduce practical empirical metrics for evaluation, and present an algorithm that optimizes these conformal objectives. Extensive experiments on vision and text benchmarks demonstrate that the proposed approach effectively removes targeted information while preserving utility.", "date": "2025-08-05", "categories": []} {"id": "2405.16828", "title": "Kernel-based Optimally Weighted Conformal Time-Series Prediction", "abstract": "In this work, we present a novel conformal prediction method for time-series, which we call Kernel-based Optimally Weighted Conformal Prediction Intervals (KOWCPI). Specifically, KOWCPI adapts the classic Reweighted Nadaraya-Watson (RNW) estimator for quantile regression on dependent data and learns optimal data-adaptive weights. Theoretically, we tackle the challenge of establishing a conditional coverage guarantee for non-exchangeable data under strong mixing conditions on the non-conformity scores. We demonstrate the superior performance of KOWCPI on real and synthetic time-series data against state-of-the-art methods, where KOWCPI achieves narrower confidence intervals without losing coverage.", "date": "2024-05-27", "categories": []} {"id": "2511.18141", "title": "Conformal Prediction for Compositional Data", "abstract": "Dirichlet regression models are suitable for compositional data, in which the response variable represents proportions that sum to one. However, there are still no well-established methods for constructing valid prediction sets in this context, especially considering the geometry of the compositional space. In this work, we investigate conformal prediction-based strategies for constructing valid predictive regions in Dirichlet regression models. We evaluate three distinct approaches: a method based on quantile residuals, an approximate construction of highest density regions (HDR), and an adaptation of the approximate HDR using grid-based discretization over the simplex. The performance of the methods was analyzed through simulation studies under different scenarios, varying the model complexity, response dimensionality, and covariate structure. The results indicated that the HDR approximation approach exhibits good robustness in terms of coverage, while the grid discretization proved effective in reducing overcoverage and the area of the prediction region compared to the original method. The quantile method provided larger prediction regions compared to the grid method, while maintaining adequate coverage. The methodologies were also applied to two real datasets: one concerning sleep stages and another on biomass allocation in plants. In both cases, the proposed methods demonstrated practical feasibility and produced coherent interpretations within the compositional space. Finally, we discuss possible extensions of this work", "date": "2025-11-22", "categories": []} {"id": "2602.10045", "title": "Conformal Prediction Sets for Instance Segmentation", "abstract": "Current instance segmentation models achieve high performance on average predictions, but lack principled uncertainty quantification: their outputs are not calibrated, and there is no guarantee that a predicted mask is close to the ground truth. To address this limitation, we introduce a conformal prediction algorithm to generate adaptive confidence sets for instance segmentation. Given an image and a pixel coordinate query, our algorithm generates a confidence set of instance predictions for that pixel, with a provable guarantee for the probability that at least one of the predictions has high Intersection-Over-Union (IoU) with the true object instance mask. We apply our algorithm to instance segmentation examples in agricultural field delineation, cell segmentation, and vehicle detection. Empirically, we find that our prediction sets vary in size based on query difficulty and attain the target coverage, outperforming existing baselines such as Learn Then Test, Conformal Risk Control, and morphological dilation-based methods. We provide versions of the algorithm with asymptotic and finite sample guarantees.", "date": "2026-02-10", "categories": []} {"id": "2602.00989", "title": "Optimal Decision-Making Based on Prediction Sets", "abstract": "Prediction sets can wrap around any ML model to cover unknown test outcomes with a guaranteed probability. Yet, it remains unclear how to use them optimally for downstream decision-making. Here, we propose a decision-theoretic framework that seeks to minimize the expected loss (risk) against a worst-case distribution consistent with the prediction set's coverage guarantee. We first characterize the minimax optimal policy for a fixed prediction set, showing that it balances the worst-case loss inside the set with a penalty for potential losses outside the set. Building on this, we derive the optimal prediction set construction that minimizes the resulting robust risk subject to a coverage constraint. Finally, we introduce Risk-Optimal Conformal Prediction (ROCP), a practical algorithm that targets these risk-minimizing sets while maintaining finite-sample distribution-free marginal coverage. Empirical evaluations on medical diagnosis and safety-critical decision-making tasks demonstrate that ROCP reduces critical mistakes compared to baselines, particularly when out-of-set errors are costly.", "date": "2026-02-01", "categories": []} {"id": "2405.16594", "title": "Training-Conditional Coverage Bounds under Covariate Shift", "abstract": "Conformal prediction methodology has recently been extended to the covariate shift setting, where the distribution of covariates differs between training and test data. While existing results ensure that the prediction sets from these methods achieve marginal coverage above a nominal level, their coverage rate conditional on the training dataset (referred to as training-conditional coverage) remains unexplored. In this paper, we address this gap by deriving upper bounds on the tail of the training-conditional coverage distribution, offering probably approximately correct (PAC) guarantees for these methods. Our results characterize the reliability of the prediction sets in terms of the severity of distributional changes and the size of the training dataset.", "date": "2024-05-26", "categories": []} {"id": "2602.06297", "title": "Time-uniform conformal and PAC prediction", "abstract": "Given that machine learning algorithms are increasingly being deployed to aid in high stakes decision-making, uncertainty quantification methods that wrap around these black box models such as conformal prediction have received much attention in recent years. In sequential settings, where data are observed/generated in a streaming fashion, traditional conformal methods do not provide any guarantee without fixing the sample size. More importantly, traditional conformal methods cannot cope with sequentially updated predictions. As such, we develop an extension of the conformal prediction and related probably approximately correct (PAC) prediction frameworks to sequential settings where the number of data points is not fixed in advance. The resulting prediction sets are anytime-valid in that their expected coverage is at the required level at any time chosen by the analyst even if this choice depends on the data. We present theoretical guarantees for our proposed methods and demonstrate their validity and utility on simulated and real datasets.", "date": "2026-02-06", "categories": []} {"id": "2509.24095", "title": "Singleton-Optimized Conformal Prediction", "abstract": "Conformal prediction can be used to construct prediction sets that cover the true outcome with a desired probability, but can sometimes lead to large prediction sets that are costly in practice. The most useful outcome is a singleton prediction-an unambiguous decision-yet existing efficiency-oriented methods primarily optimize average set size. Motivated by this, we propose a new nonconformity score that aims to minimize the probability of producing non-singleton sets. Starting from a non-convex constrained optimization problem as a motivation, we provide a geometric reformulation and associated algorithm for computing the nonconformity score and associated split conformal prediction sets in O(K) time for K-class problems. Using this score in split conformal prediction leads to our proposed Singleton-Optimized Conformal Prediction (SOCOP) method. We evaluate our method in experiments on image classification and LLM multiple-choice question-answering, comparing with standard nonconformity scores such as the (negative) label probability estimates and their cumulative distribution function; both of which are motivated by optimizing length. The results show that SOCOP increases singleton frequency (sometimes by over 20%) compared to the above scores, with minimal impact on average set size.", "date": "2025-09-28", "categories": []} {"id": "2511.05610", "title": "Conformal Prediction-Driven Adaptive Sampling for Digital Water Twins", "abstract": "Digital Twins (DTs) for Water Distribution Networks (WDNs) require accurate state estimation with limited sensors. Uniform sampling often wastes resources across nodes with different uncertainty. We propose an adaptive framework combining LSTM forecasting and Conformal Prediction (CP) to estimate node-wise uncertainty and focus sensing on the most uncertain points. Marginal CP is used for its low computational cost, suitable for real-time DTs. Experiments on Hanoi, Net3, and CTOWN show 33--34\\% lower demand error than uniform sampling at 40\\% coverage and maintain 89.4--90.2\\% empirical coverage with only 5--10\\% extra computation.", "date": "2025-11-06", "categories": []} {"id": "2602.04364", "title": "Anytime-Valid Conformal Risk Control", "abstract": "Prediction sets provide a means of quantifying the uncertainty in predictive tasks. Using held out calibration data, conformal prediction and risk control can produce prediction sets that exhibit statistically valid error control in a computationally efficient manner. However, in the standard formulations, the error is only controlled on average over many possible calibration datasets of fixed size. In this paper, we extend the control to remain valid with high probability over a cumulatively growing calibration dataset at any time point. We derive such guarantees using quantile-based arguments and illustrate the applicability of the proposed framework to settings involving distribution shift. We further establish a matching lower bound and show that our guarantees are asymptotically tight. Finally, we demonstrate the practical performance of our methods through both simulations and real-world numerical examples.", "date": "2026-02-04", "categories": []} {"id": "2407.03094", "title": "Conformal Prediction for Causal Effects of Continuous Treatments", "abstract": "Uncertainty quantification of causal effects is crucial for safety-critical applications such as personalized medicine. A powerful approach for this is conformal prediction, which has several practical benefits due to model-agnostic finite-sample guarantees. Yet, existing methods for conformal prediction of causal effects are limited to binary/discrete treatments and make highly restrictive assumptions such as known propensity scores. In this work, we provide a novel conformal prediction method for potential outcomes of continuous treatments. We account for the additional uncertainty introduced through propensity estimation so that our conformal prediction intervals are valid even if the propensity score is unknown. Our contributions are three-fold: (1) We derive finite-sample prediction intervals for potential outcomes of continuous treatments. (2) We provide an algorithm for calculating the derived intervals. (3) We demonstrate the effectiveness of the conformal prediction intervals in experiments on synthetic and real-world datasets. To the best of our knowledge, we are the first to propose conformal prediction for continuous treatments when the propensity score is unknown and must be estimated from data.", "date": "2024-07-03", "categories": []} {"id": "2602.02583", "title": "Copula-Based Aggregation and Context-Aware Conformal Prediction for Reliable Renewable Energy Forecasting", "abstract": "The rapid growth of renewable energy penetration has intensified the need for reliable probabilistic forecasts to support grid operations at aggregated (fleet or system) levels. In practice, however, system operators often lack access to fleet-level probabilistic models and instead rely on site-level forecasts produced by heterogeneous third-party providers. Constructing coherent and calibrated fleet-level probabilistic forecasts from such inputs remains challenging due to complex cross-site dependencies and aggregation-induced miscalibration. This paper proposes a calibrated probabilistic aggregation framework that directly converts site-level probabilistic forecasts into reliable fleet-level forecasts in settings where system-level models cannot be trained or maintained. The framework integrates copula-based dependence modeling to capture cross-site correlations with Context-Aware Conformal Prediction (CACP) to correct miscalibration at the aggregated level. This combination enables dependence-aware aggregation while providing valid coverage and maintaining sharp prediction intervals. Experiments on large-scale solar generation datasets from MISO, ERCOT, and SPP demonstrate that the proposed Copula+CACP approach consistently achieves near-nominal coverage with significantly sharper intervals than uncalibrated aggregation baselines.", "date": "2026-01-31", "categories": []} {"id": "2602.03168", "title": "Online Conformal Prediction via Universal Portfolio Algorithms", "abstract": "Online conformal prediction (OCP) seeks prediction intervals that achieve long-run $1-\\alpha$ coverage for arbitrary (possibly adversarial) data streams, while remaining as informative as possible. Existing OCP methods often require manual learning-rate tuning to work well, and may also require algorithm-specific analyses. Here, we develop a general regret-to-coverage theory for interval-valued OCP based on the $(1-\\alpha)$-pinball loss. Our first contribution is to identify \\emph{linearized regret} as a key notion, showing that controlling it implies coverage bounds for any online algorithm. This relies on a black-box reduction that depends only on the Fenchel conjugate of an upper bound on the linearized regret. Building on this theory, we propose UP-OCP, a parameter-free method for OCP, via a reduction to a two-asset portfolio selection problem, leveraging universal portfolio algorithms. We show strong finite-time bounds on the miscoverage of UP-OCP, even for polynomially growing predictions. Extensive experiments support that UP-OCP delivers consistently better size/coverage trade-offs than prior online conformal baselines.", "date": "2026-02-03", "categories": []} {"id": "2602.03799", "title": "Conformal Reachability for Safe Control in Unknown Environments", "abstract": "Designing provably safe control is a core problem in trustworthy autonomy. However, most prior work in this regard assumes either that the system dynamics are known or deterministic, or that the state and action space are finite, significantly limiting application scope. We address this limitation by developing a probabilistic verification framework for unknown dynamical systems which combines conformal prediction with reachability analysis. In particular, we use conformal prediction to obtain valid uncertainty intervals for the unknown dynamics at each time step, with reachability then verifying whether safety is maintained within the conformal uncertainty bounds. Next, we develop an algorithmic approach for training control policies that optimize nominal reward while also maximizing the planning horizon with sound probabilistic safety guarantees. We evaluate the proposed approach in seven safe control settings spanning four domains -- cartpole, lane following, drone control, and safe navigation -- for both affine and nonlinear safety specifications. Our experiments show that the policies we learn achieve the strongest provable safety guarantees while still maintaining high average reward.", "date": "2026-02-03", "categories": []} {"id": "2409.20534", "title": "End-to-End Conformal Calibration for Optimization Under Uncertainty", "abstract": "Machine learning can significantly improve performance for decision-making under uncertainty across a wide range of domains. However, ensuring robustness guarantees requires well-calibrated uncertainty estimates, which can be difficult to achieve with neural networks. Moreover, in high-dimensional settings, there may be many valid uncertainty estimates, each with its own performance profile - i.e., not all uncertainty is equally valuable for downstream decision-making. To address this problem, this paper develops an end-to-end framework to learn uncertainty sets for conditional robust optimization in a way that is informed by the downstream decision-making loss, with robustness and calibration guarantees provided by conformal prediction. In addition, we propose to represent general convex uncertainty sets with partially input-convex neural networks, which are learned as part of our framework. Our approach consistently improves upon two-stage estimate-then-optimize baselines on concrete applications in energy storage arbitrage and portfolio optimization.", "date": "2024-09-30", "categories": []} {"id": "2509.16324", "title": "Auto-bidding under Return-on-Spend Constraints with Uncertainty Quantification", "abstract": "Auto-bidding systems are widely used in advertising to automatically determine bid values under constraints such as total budget and Return-on-Spend (RoS) targets. Existing works often assume that the value of an ad impression, such as the conversion rate, is known. This paper considers the more realistic scenario where the true value is unknown. We propose a novel method that uses conformal prediction to quantify the uncertainty of these values based on machine learning methods trained on historical bidding data with contextual features, without assuming the data are i.i.d. This approach is compatible with current industry systems that use machine learning to predict values. Building on prediction intervals, we introduce an adjusted value estimator derived from machine learning predictions, and show that it provides performance guarantees without requiring knowledge of the true value. We apply this method to enhance existing auto-bidding algorithms with budget and RoS constraints, and establish theoretical guarantees for achieving high reward while keeping RoS violations low. Empirical results on both simulated and real-world industrial datasets demonstrate that our approach improves performance while maintaining computational efficiency.", "date": "2025-09-19", "categories": []} {"id": "2602.01285", "title": "Multi-LLM Adaptive Conformal Inference for Reliable LLM Responses", "abstract": "Ensuring factuality is essential for the safe use of Large Language Models (LLMs) in high-stakes domains such as medicine and law. Conformal inference provides distribution-free guarantees, but existing approaches are either overly conservative, discarding many true-claims, or rely on adaptive error rates and simple linear models that fail to capture complex group structures. To address these challenges, we reformulate conformal inference in a multiplicative filtering setting, modeling factuality as a product of claim-level scores. Our method, Multi-LLM Adaptive Conformal Inference (MACI), leverages ensembles to produce more accurate factuality-scores, which in our experiments led to higher retention, while validity is preserved through group-conditional calibration. Experiments show that MACI consistently achieves user-specified coverage with substantially higher retention and lower time cost than baselines. Our repository is available at https://github.com/MLAI-Yonsei/MACI", "date": "2026-02-01", "categories": []} {"id": "2602.01667", "title": "Quantifying Epistemic Predictive Uncertainty in Conformal Prediction", "abstract": "We study the problem of quantifying epistemic predictive uncertainty (EPU) -- that is, uncertainty faced at prediction time due to the existence of multiple plausible predictive models -- within the framework of conformal prediction (CP). To expose the implicit model multiplicity underlying CP, we build on recent results showing that, under a mild assumption, any full CP procedure induces a set of closed and convex predictive distributions, commonly referred to as a credal set. Importantly, the conformal prediction region (CPR) coincides exactly with the set of labels to which all distributions in the induced credal set assign probability at least $1-\\alpha$. As our first contribution, we prove that this characterisation also holds in split CP. Building on this connection, we then propose a computationally efficient and analytically tractable uncertainty measure, based on \\emph{Maximum Mean Imprecision}, to quantify the EPU by measuring the degree of conflicting information within the induced credal set. Experiments on active learning and selective classification demonstrate that the quantified EPU provides substantially more informative and fine-grained uncertainty assessments than reliance on CPR size alone. More broadly, this work highlights the potential of CP serving as a principled basis for decision-making under epistemic uncertainty.", "date": "2026-02-02", "categories": []} {"id": "2602.01912", "title": "Reliable Real-Time Value at Risk Estimation via Quantile Regression Forest with Conformal Calibration", "abstract": "Rapidly evolving market conditions call for real-time risk monitoring, but its online estimation remains challenging. In this paper, we study the online estimation of one of the most widely used risk measures, Value at Risk (VaR). Its accurate and reliable estimation is essential for timely risk control and informed decision-making. We propose to use the quantile regression forest in the offline-simulation-online-estimation (OSOA) framework. Specifically, the quantile regression forest is trained offline to learn the relationship between the online VaR and risk factors, and real-time VaR estimates are then produced online by incorporating observed risk factors. To further ensure reliability, we develop a conformalized estimator that calibrates the online VaR estimates. To the best of our knowledge, we are the first to leverage conformal calibration to estimate real-time VaR reliably based on the OSOA formulation. Theoretical analysis establishes the consistency and coverage validity of the proposed estimators. Numerical experiments confirm the proposed method and demonstrate its effectiveness in practice.", "date": "2026-02-02", "categories": []} {"id": "2402.12683", "title": "TorchCP: A Python Library for Conformal Prediction", "abstract": "Conformal prediction (CP) is a powerful statistical framework that generates prediction intervals or sets with guaranteed coverage probability. While CP algorithms have evolved beyond traditional classifiers and regressors to sophisticated deep learning models like deep neural networks (DNNs), graph neural networks (GNNs), and large language models (LLMs), existing CP libraries often lack the model support and scalability for large-scale deep learning (DL) scenarios. This paper introduces TorchCP, a PyTorch-native library designed to integrate state-of-the-art CP algorithms into DL techniques, including DNN-based classifiers/regressors, GNNs, and LLMs. Released under the LGPL-3.0 license, TorchCP comprises about 16k lines of code, validated with 100\\% unit test coverage and detailed documentation. Notably, TorchCP enables CP-specific training algorithms, online prediction, and GPU-accelerated batch processing, achieving up to 90\\% reduction in inference time on large datasets. With its low-coupling design, comprehensive suite of advanced methods, and full GPU scalability, TorchCP empowers researchers and practitioners to enhance uncertainty quantification across cutting-edge applications.", "date": "2024-02-20", "categories": []} {"id": "2509.21479", "title": "Filtering with Confidence: When Data Augmentation Meets Conformal Prediction", "abstract": "With promising empirical performance across a wide range of applications, synthetic data augmentation appears a viable solution to data scarcity and the demands of increasingly data-intensive models. Its effectiveness lies in expanding the training set in a way that reduces estimator variance while introducing only minimal bias. Controlling this bias is therefore critical: effective data augmentation should generate diverse samples from the same underlying distribution as the training set, with minimal shifts. In this paper, we propose conformal data augmentation, a principled data filtering framework that leverages the power of conformal prediction to produce diverse synthetic data while filtering out poor-quality generations with provable risk control. Our method is simple to implement, requires no access to internal model logits, nor large-scale model retraining. We demonstrate the effectiveness of our approach across multiple tasks, including topic prediction, sentiment analysis, image classification, and fraud detection, showing consistent performance improvements of up to 40 percentage points (pp) in $F_1$ score over unaugmented baselines, and 4~pp over other filtered augmentation baselines.", "date": "2025-09-25", "categories": []} {"id": "2601.05219", "title": "CAOS: Conformal Aggregation of One-Shot Predictors", "abstract": "One-shot prediction enables rapid adaptation of pretrained foundation models to new tasks using only one labeled example, but lacks principled uncertainty quantification. While conformal prediction provides finite-sample coverage guarantees, standard split conformal methods are inefficient in the one-shot setting due to data splitting and reliance on a single predictor. We propose Conformal Aggregation of One-Shot Predictors (CAOS), a conformal framework that adaptively aggregates multiple one-shot predictors and uses a leave-one-out calibration scheme to fully exploit scarce labeled data. Despite violating classical exchangeability assumptions, we prove that CAOS achieves valid marginal coverage using a monotonicity-based argument. Experiments on one-shot facial landmarking and RAFT text classification tasks show that CAOS produces substantially smaller prediction sets than split conformal baselines while maintaining reliable coverage.", "date": "2026-01-08", "categories": []} {"id": "2601.18509", "title": "Conformal Prediction Algorithms for Time Series Forecasting: Methods and Benchmarking", "abstract": "Reliable uncertainty quantification is of critical importance in time series forecasting, yet traditional methods often rely on restrictive distributional assumptions. Conformal prediction (CP) has emerged as a promising distribution-free framework for generating prediction intervals with rigorous theoretical guarantees. However, applying CP to sequential data presents a primary challenge: the temporal dependencies inherent in time series fundamentally violate the core assumption of data exchangeability, upon which standard CP guarantees are built. This paper critically examines the main categories of algorithmic solutions designed to address this conflict. We survey and benchmark methods that relax the exchangeability assumption, those that redefine the data unit to be a collection of independent time series, approaches that explicitly model the dynamics of the prediction residuals, and online learning algorithms that adapt to distribution shifts to maintain long-run coverage. We use AutoARIMA as the base forecaster on a large-scale monthly sales dataset, evaluating marginal coverage, interval width, and the Winkler score. Our benchmark results show that multi-step split conformal prediction method meets the 90% coverage threshold and demonstrates the best efficiency.", "date": "2026-01-26", "categories": []} {"id": "2601.22298", "title": "Conformal Prediction for Generative Models via Adaptive Cluster-Based Density Estimation", "abstract": "Conditional generative models map input variables to complex, high-dimensional distributions, enabling realistic sample generation in a diverse set of domains. A critical challenge with these models is the absence of calibrated uncertainty, which undermines trust in individual outputs for high-stakes applications. To address this issue, we propose a systematic conformal prediction approach tailored to conditional generative models, leveraging density estimation on model-generated samples. We introduce a novel method called CP4Gen, which utilizes clustering-based density estimation to construct prediction sets that are less sensitive to outliers, more interpretable, and of lower structural complexity than existing methods. Extensive experiments on synthetic datasets and real-world applications, including climate emulation tasks, demonstrate that CP4Gen consistently achieves superior performance in terms of prediction set volume and structural simplicity. Our approach offers practitioners a powerful tool for uncertainty estimation associated with conditional generative models, particularly in scenarios demanding rigorous and interpretable prediction sets.", "date": "2026-01-29", "categories": []} {"id": "2601.23128", "title": "Distribution-informed Efficient Conformal Prediction for Full Ranking", "abstract": "Quantifying uncertainty is critical for the safe deployment of ranking models in real-world applications. Recent work offers a rigorous solution using conformal prediction in a full ranking scenario, which aims to construct prediction sets for the absolute ranks of test items based on the relative ranks of calibration items. However, relying on upper bounds of non-conformity scores renders the method overly conservative, resulting in substantially large prediction sets. To address this, we propose Distribution-informed Conformal Ranking (DCR), which produces efficient prediction sets by deriving the exact distribution of non-conformity scores. In particular, we find that the absolute ranks of calibration items follow Negative Hypergeometric distributions, conditional on their relative ranks. DCR thus uses the rank distribution to derive non-conformity score distribution and determine conformal thresholds. We provide theoretical guarantees that DCR achieves improved efficiency over the baseline while ensuring valid coverage under mild assumptions. Extensive experiments demonstrate the superiority of DCR, reducing average prediction set size by up to 36%, while maintaining valid coverage.", "date": "2026-01-30", "categories": []} {"id": "2506.05544", "title": "Online Conformal Model Selection for Nonstationary Time Series", "abstract": "This paper introduces the MPS (Model Prediction Set), a novel framework for online model selection for nonstationary time series. Classical model selection methods, such as information criteria and cross-validation, rely heavily on the stationarity assumption and often fail in dynamic environments which undergo gradual or abrupt changes over time. Yet real-world data are rarely stationary, and model selection under nonstationarity remains a largely open problem. To tackle this challenge, we combine conformal inference with model confidence sets to develop a procedure that adaptively selects models best suited to the evolving dynamics at any given time. Concretely, the MPS updates in real time a confidence set of candidate models that covers the best model for the next time period with a specified long-run probability, while adapting to nonstationarity of unknown forms. Through simulations and real-world data analysis, we demonstrate that MPS reliably and efficiently identifies optimal models under nonstationarity, an essential capability lacking in offline methods. Moreover, MPS frequently produces high-quality sets with small cardinality, whose evolution offers deeper insights into changing dynamics. As a generic framework, MPS accommodates any data-generating process, data structure, model class, training method, and evaluation metric, making it broadly applicable across diverse problem settings.", "date": "2025-06-05", "categories": []} {"id": "2601.18997", "title": "Anatomically-aware conformal prediction for medical image segmentation with random walks", "abstract": "The reliable deployment of deep learning in medical imaging requires uncertainty quantification that provides rigorous error guarantees while remaining anatomically meaningful. Conformal prediction (CP) is a powerful distribution-free framework for constructing statistically valid prediction intervals. However, standard applications in segmentation often ignore anatomical context, resulting in fragmented, spatially incoherent, and over-segmented prediction sets that limit clinical utility. To bridge this gap, this paper proposes Random-Walk Conformal Prediction (RW-CP), a model-agnostic framework which can be added on top of any segmentation method. RW-CP enforces spatial coherence to generate anatomically valid sets. Our method constructs a k-nearest neighbour graph from pre-trained vision foundation model features and applies a random walk to diffuse uncertainty. The random walk diffusion regularizes the non-conformity scores, making the prediction sets less sensitive to the conformal calibration parameter $\\lambda$, ensuring more stable and continuous anatomical boundaries. RW-CP maintains rigorous marginal coverage while significantly improving segmentation quality. Evaluations on multi-modal public datasets show improvements of up to $35.4\\%$ compared to standard CP baselines, given an allowable error rate of $\\alpha=0.1$.", "date": "2026-01-26", "categories": []} {"id": "2410.05263", "title": "Bias-Aware Conformal Prediction for Metric-Based Imaging Pipelines", "abstract": "Reliable confidence measures of metrics derived from medical imaging reconstruction pipelines would improve the standard of decision-making in many clinical workflows. Conformal Prediction (CP) provides a robust framework for producing calibrated prediction intervals, but standard CP formulations face a critical challenge in the imaging pipeline: common mismatches between image reconstruction objectives and downstream metrics can introduce systematic prediction deviations from ground truth values, known as bias. These biases in turn compromise the efficiency of prediction intervals, which is a problem that has been unexplored in the CP literature. In this study, we formalize the behavior of symmetric (where bounds expand equally in both directions) and asymmetric (where bounds expand unequally) formulations for common non-conformity scores in CP in the presence of bias, and argue that this measurable bias must inform the choice of CP formulation. We theoretically and empirically demonstrate that symmetric intervals are inflated by a factor of two times the magnitude of bias while asymmetric intervals remain unaffected by bias, and provide conditions under which each formulation produces tighter intervals. We empirically validated our theoretical analyses on sparse-view CT reconstruction for downstream radiotherapy planning. Our work enables users of medical imaging pipelines to proactively select optimal CP formulations, thereby improving interval length efficiency for critical downstream metrics.", "date": "2024-10-07", "categories": []} {"id": "2601.13102", "title": "Approximate full conformal prediction in an RKHS", "abstract": "Full conformal prediction is a framework that implicitly formulates distribution-free confidence prediction regions for a wide range of estimators. However, a classical limitation of the full conformal framework is the computation of the confidence prediction regions, which is usually impossible since it requires training infinitely many estimators (for real-valued prediction for instance). The main purpose of the present work is to describe a generic strategy for designing a tight approximation to the full conformal prediction region that can be efficiently computed. Along with this approximate confidence region, a theoretical quantification of the tightness of this approximation is developed, depending on the smoothness assumptions on the loss and score functions. The new notion of thickness is introduced for quantifying the discrepancy between the approximate confidence region and the full conformal one.", "date": "2026-01-19", "categories": []} {"id": "2507.05470", "title": "Temporal Conformal Prediction (TCP): A Distribution-Free Statistical and Machine Learning Framework for Adaptive Risk Forecasting", "abstract": "We propose \\textbf{Temporal Conformal Prediction (TCP)}, a distribution-free framework for constructing well-calibrated prediction intervals in nonstationary time series. TCP couples a modern quantile forecaster with a rolling split-conformal calibration layer; its \\textbf{TCP-RM} variant adds an online Robbins-Monro offset to steer coverage in real time. We benchmark TCP against GARCH, Historical Simulation, Quantile Regression (QR), linear QR, and Adaptive Conformal Inference (ACI) across S\\&P 500, Bitcoin, and Gold. Three results are consistent. First, QR baselines yield the sharpest intervals but are materially under-calibrated; even ACI remains below the 95\\% target. Second, TCP achieves near-nominal coverage, yielding intervals slightly wider than Historical Simulation (e.g., S\\&P 500: 5.21 vs.\\ 5.06). Third, the RM update changes calibration only marginally at default hyperparameters. Crisis-window visualizations (March 2020) show TCP promptly expanding and contracting intervals as volatility spikes. A sensitivity study confirms robustness to hyperparameters. Overall, TCP bridges statistical inference and machine learning, providing a practical solution for calibrated risk forecasting under distribution shift.", "date": "2025-07-07", "categories": []} {"id": "2601.01147", "title": "Conformal Blindness: A Note on $A$-Cryptic change-points", "abstract": "Conformal Test Martingales (CTMs) are a standard method within the Conformal Prediction framework for testing the crucial assumption of data exchangeability by monitoring deviations from uniformity in the p-value sequence. Although exchangeability implies uniform p-values, the converse does not hold. This raises the question of whether a significant break in exchangeability can occur, such that the p-values remain uniform, rendering CTMs blind. We answer this affirmatively, demonstrating the phenomenon of \\emph{conformal blindness}.\n Through explicit construction, for the theoretically ideal ``predictive oracle'' conformity measure (given by the true conditional density), we demonstrate the possibility of an \\emph{$A$-cryptic change-point} (where $A$ refers to the conformity measure). Using bivariate Gaussian distributions, we identify a line along which a change in the marginal means does not alter the distribution of the conformity scores, thereby producing perfectly uniform p-values.\n Simulations confirm that even a massive distribution shift can be perfectly cryptic to the CTM, highlighting a fundamental limitation and emphasising the critical role of the alignment of the conformity measure with potential shifts.\n By contrasting the predictive oracle with recent results on detection-optimal scores, we emphasise that validity monitoring in safety-critical systems requires careful separation of predictive and diagnostic goals.", "date": "2026-01-03", "categories": []} {"id": "2601.09522", "title": "Class Adaptive Conformal Training", "abstract": "Deep neural networks have achieved remarkable success across a variety of tasks, yet they often suffer from unreliable probability estimates. As a result, they can be overconfident in their predictions. Conformal Prediction (CP) offers a principled framework for uncertainty quantification, yielding prediction sets with rigorous coverage guarantees. Existing conformal training methods optimize for overall set size, but shaping the prediction sets in a class-conditional manner is not straightforward and typically requires prior knowledge of the data distribution. In this work, we introduce Class Adaptive Conformal Training (CaCT), which formulates conformal training as an augmented Lagrangian optimization problem that adaptively learns to shape prediction sets class-conditionally without making any distributional assumptions. Experiments on multiple benchmark datasets, including standard and long-tailed image recognition as well as text classification, demonstrate that CaCT consistently outperforms prior conformal training methods, producing significantly smaller and more informative prediction sets while maintaining the desired coverage guarantees.", "date": "2026-01-14", "categories": []} {"id": "2510.02471", "title": "Predictive inference for time series: why is split conformal effective despite temporal dependence?", "abstract": "We consider the problem of uncertainty quantification for prediction in a time series: if we use past data to forecast the next time point, can we provide valid prediction intervals around our forecasts? To avoid placing distributional assumptions on the data, in recent years the conformal prediction method has been a popular approach for predictive inference, since it provides distribution-free coverage for any iid or exchangeable data distribution. However, in the time series setting, the strong empirical performance of conformal prediction methods is not well understood, since even short-range temporal dependence is a strong violation of the exchangeability assumption. Using predictors with \"memory\" -- i.e., predictors that utilize past observations, such as autoregressive models -- further exacerbates this problem. In this work, we examine the theoretical properties of split conformal prediction in the time series setting, including the case where predictors may have memory. Our results bound the loss of coverage of these methods in terms of a new \"switch coefficient\", measuring the extent to which temporal dependence within the time series creates violations of exchangeability. Our characterization of the coverage probability is sharp over the class of stationary, $\\beta$-mixing processes. Along the way, we introduce tools that may prove useful in analyzing other predictive inference methods for dependent data.", "date": "2025-10-02", "categories": []} {"id": "2601.07201", "title": "CalPro: Prior-Aware Evidential--Conformal Prediction with Structure-Aware Guarantees for Protein Structures", "abstract": "Deep protein structure predictors such as AlphaFold provide confidence estimates (e.g., pLDDT) that are often miscalibrated and degrade under distribution shifts across experimental modalities, temporal changes, and intrinsically disordered regions. We introduce CalPro, a prior-aware evidential-conformal framework for shift-robust uncertainty quantification. CalPro combines (i) a geometric evidential head that outputs Normal-Inverse-Gamma predictive distributions via a graph-based architecture; (ii) a differentiable conformal layer that enables end-to-end training with finite-sample coverage guarantees; and (iii) domain priors (disorder, flexibility) encoded as soft constraints. We derive structure-aware coverage guarantees under distribution shift using PAC-Bayesian bounds over ambiguity sets, and show that CalPro maintains near-nominal coverage while producing tighter intervals than standard conformal methods in regions where priors are informative. Empirically, CalPro exhibits at most 5% coverage degradation across modalities (vs. 15-25% for baselines), reduces calibration error by 30-50%, and improves downstream ligand-docking success by 25%. Beyond proteins, CalPro applies to structured regression tasks in which priors encode local reliability, validated on non-biological benchmarks.", "date": "2026-01-12", "categories": []} {"id": "2409.20412", "title": "Conformal Prediction for Dose-Response Models with Continuous Treatments", "abstract": "Understanding the dose-response relation between a continuous treatment and the outcome for an individual can greatly drive decision-making, particularly in areas like personalized drug dosing and personalized healthcare interventions. Point estimates are often insufficient in these high-risk environments, highlighting the need for uncertainty quantification to support informed decisions. Conformal prediction, a distribution-free and model-agnostic method for uncertainty quantification, has seen limited application in continuous treatments or dose-response models. To address this gap, we propose a novel methodology that frames the causal dose-response problem as a covariate shift, leveraging weighted conformal prediction. By incorporating propensity estimation, conformal predictive systems, and likelihood ratios, we present a practical solution for generating prediction intervals for dose-response models. Additionally, our method approximates local coverage for every treatment value by applying kernel functions as weights in weighted conformal prediction. Finally, we use a new synthetic benchmark dataset to demonstrate the significance of covariate shift assumptions in achieving robust prediction intervals for dose-response models.", "date": "2024-09-30", "categories": []} {"id": "2601.02769", "title": "Fast Conformal Prediction using Conditional Interquantile Intervals", "abstract": "We introduce Conformal Interquantile Regression (CIR), a conformal regression method that efficiently constructs near-minimal prediction intervals with guaranteed coverage. CIR leverages black-box machine learning models to estimate outcome distributions through interquantile ranges, transforming these estimates into compact prediction intervals while achieving approximate conditional coverage. We further propose CIR+ (Conditional Interquantile Regression with More Comparison), which enhances CIR by incorporating a width-based selection rule for interquantile intervals. This refinement yields narrower prediction intervals while maintaining comparable coverage, though at the cost of slightly increased computational time. Both methods address key limitations of existing distributional conformal prediction approaches: they handle skewed distributions more effectively than Conformalized Quantile Regression, and they achieve substantially higher computational efficiency than Conformal Histogram Regression by eliminating the need for histogram construction. Extensive experiments on synthetic and real-world datasets demonstrate that our methods optimally balance predictive accuracy and computational efficiency compared to existing approaches.", "date": "2026-01-06", "categories": []} {"id": "2601.02998", "title": "Multi-Distribution Robust Conformal Prediction", "abstract": "In many fairness and distribution robustness problems, one has access to labeled data from multiple source distributions yet the test data may come from an arbitrary member or a mixture of them. We study the problem of constructing a conformal prediction set that is uniformly valid across multiple, heterogeneous distributions, in the sense that no matter which distribution the test point is from, the coverage of the prediction set is guaranteed to exceed a pre-specified level. We first propose a max-p aggregation scheme that delivers finite-sample, multi-distribution coverage given any conformity scores associated with each distribution. Upon studying several efficiency optimization programs subject to uniform coverage, we prove the optimality and tightness of our aggregation scheme, and propose a general algorithm to learn conformity scores that lead to efficient prediction sets after the aggregation under standard conditions. We discuss how our framework relates to group-wise distributionally robust optimization, sub-population shift, fairness, and multi-source learning. In synthetic and real-data experiments, our method delivers valid worst-case coverage across multiple distributions while greatly reducing the set size compared with naively applying max-p aggregation to single-source conformity scores, and can be comparable in size to single-source prediction sets with popular, standard conformity scores.", "date": "2026-01-06", "categories": []} {"id": "2408.09881", "title": "Uncertainty Quantification of Surrogate Models using Conformal Prediction", "abstract": "Data-driven surrogate models offer quick approximations to complex numerical and experimental systems but typically lack uncertainty quantification, limiting their reliability in safety-critical applications. While Bayesian methods provide uncertainty estimates, they offer no statistical guarantees and struggle with high-dimensional spatio-temporal problems due to computational costs. We present a conformal prediction (CP) framework that provides statistically guaranteed marginal coverage for surrogate models in a model-agnostic manner with near-zero computational cost. Our approach handles high-dimensional spatio-temporal outputs by performing cell-wise calibration while preserving the tensorial structure of predictions. Through extensive empirical evaluation across diverse applications including fluid dynamics, magnetohydrodynamics, weather forecasting, and fusion diagnostics, we demonstrate that CP achieves empirical coverage with valid error bars regardless of model architecture, training regime, or output dimensionality. We evaluate three nonconformity scores (conformalised quantile regression, absolute error residual, and standard deviation) for both deterministic and probabilistic models, showing that guaranteed coverage holds even for out-of-distribution predictions where models are deployed on physics regimes different from training data. Calibration requires only seconds to minutes on standard hardware. The framework enables rigorous validation of pre-trained surrogate models for downstream applications without retraining. While CP provides marginal rather than conditional coverage and assumes exchangeability between calibration and test data, our method circumvents the curse of dimensionality inherent in traditional uncertainty quantification approaches, offering a practical tool for trustworthy deployment of machine learning in physical sciences.", "date": "2024-08-19", "categories": []} {"id": "2601.00908", "title": "Conformal Prediction Under Distribution Shift: A COVID-19 Natural Experiment", "abstract": "Conformal prediction guarantees degrade under distribution shift. We study this using COVID-19 as a natural experiment across 8 supply chain tasks. Despite identical severe feature turnover (Jaccard approximately 0), coverage drops vary from 0% to 86.7%, spanning two orders of magnitude. Using SHapley Additive exPlanations (SHAP) analysis, we find catastrophic failures correlate with single-feature dependence (rho = 0.714, p = 0.047). Catastrophic tasks concentrate importance in one feature (4.5x increase), while robust tasks redistribute across many (10-20x). Quarterly retraining restores catastrophic task coverage from 22% to 41% (+19 pp, p = 0.04), but provides no benefit for robust tasks (99.8% coverage). Exploratory analysis of 4 additional tasks with moderate feature stability (Jaccard 0.13-0.86) reveals feature stability, not concentration, determines robustness, suggesting concentration effects apply specifically to severe shifts. We provide a decision framework: monitor SHAP concentration before deployment; retrain quarterly if vulnerable (>40% concentration); skip retraining if robust.", "date": "2026-01-01", "categories": []} {"id": "2601.01223", "title": "Adaptive Conformal Prediction via Bayesian Uncertainty Weighting for Hierarchical Healthcare Data", "abstract": "Clinical decision-making demands uncertainty quantification that provides both distribution-free coverage guarantees and risk-adaptive precision, requirements that existing methods fail to jointly satisfy. We present a hybrid Bayesian-conformal framework that addresses this fundamental limitation in healthcare predictions. Our approach integrates Bayesian hierarchical random forests with group-aware conformal calibration, using posterior uncertainties to weight conformity scores while maintaining rigorous coverage validity. Evaluated on 61,538 admissions across 3,793 U.S. hospitals and 4 regions, our method achieves target coverage (94.3% vs 95% target) with adaptive precision: 21% narrower intervals for low-uncertainty cases while appropriately widening for high-risk predictions. Critically, we demonstrate that well-calibrated Bayesian uncertainties alone severely under-cover (14.1%), highlighting the necessity of our hybrid approach. This framework enables risk-stratified clinical protocols, efficient resource planning for high-confidence predictions, and conservative allocation with enhanced oversight for uncertain cases, providing uncertainty-aware decision support across diverse healthcare settings.", "date": "2026-01-03", "categories": []} {"id": "2601.01692", "title": "Enhanced Multi-model Online Conformal Prediction", "abstract": "Conformal prediction is a framework for uncertainty quantification that constructs prediction sets for previously unseen data, guaranteeing coverage of the true label with a specified probability. However, the efficiency of these prediction sets, measured by their size, depends on the choice of the underlying learning model. Relying on a single fixed model may lead to suboptimal performance in online environments, as a single model may not consistently perform well across all time steps. To mitigate this, prior work has explored selecting a model from a set of candidates. However, this approach becomes computationally expensive as the number of candidate models increases. Moreover, poorly performing models in the set may also hinder the effectiveness. To tackle this challenge, this work develops a novel multi-model online conformal prediction algorithm that reduces computational complexity and improves prediction efficiency. At each time step, a bipartite graph is generated to identify a subset of effective models, from which a model is selected to construct the prediction set. Experiments demonstrate that our method outperforms existing multi-model conformal prediction techniques in terms of both prediction set size and computational efficiency.", "date": "2026-01-04", "categories": []} {"id": "2501.14570", "title": "coverforest: Conformal Predictions with Random Forest in Python", "abstract": "Conformal prediction provides a framework for uncertainty quantification, specifically in the forms of prediction intervals and sets with distribution-free guaranteed coverage. While recent cross-conformal techniques such as CV+ and Jackknife+-after-bootstrap achieve better data efficiency than traditional split conformal methods, they incur substantial computational costs due to required pairwise comparisons between training and test samples' out-of-bag scores. Observing that these methods naturally extend from ensemble models, particularly random forests, we leverage existing optimized random forest implementations to enable efficient cross-conformal predictions.\n We present coverforest, a Python package that implements efficient conformal prediction methods specifically optimized for random forests. coverforest supports both regression and classification tasks through various conformal prediction methods, including split conformal, CV+, Jackknife+-after-bootstrap, and adaptive prediction sets. Our package leverages parallel computing and Cython optimizations to speed up out-of-bag calculations. Our experiments demonstrate that coverforest's predictions achieve the desired level of coverage. In addition, its training and prediction times can be faster than an existing implementation by 2--9 times. The source code for the coverforest is hosted on GitHub at https://github.com/donlap/coverforest.", "date": "2025-01-24", "categories": []} {"id": "2512.22682", "title": "Conformal Prediction Sets for Next-Token Prediction in Large Language Models: Balancing Coverage Guarantees with Set Efficiency", "abstract": "Deploying large language models (LLMs) in high-stakes domains requires rigorous uncertainty quantification, yet standard softmax probabilities are often poorly calibrated. We present a systematic study of Adaptive Prediction Sets (APS) applied to next-token prediction in transformer-based models with large vocabularies (greater than 250,000 tokens). Our central contribution is the identification of a coverage-efficiency tradeoff: while naive conformal prediction achieves valid coverage, it produces prediction sets of hundreds of tokens, rendering them uninformative. We propose Vocabulary-Aware Conformal Prediction (VACP), a framework that leverages semantic masking and temperature-adjusted scoring to reduce the effective prediction space while provably maintaining marginal coverage. Experiments on Gemma-2B using SQUAD and WikiText benchmarks demonstrate that VACP achieves 89.7 percent empirical coverage (90 percent target) while reducing the mean prediction set size from 847 tokens to 4.3 tokens -- a 197x improvement in efficiency. We provide a theoretical analysis of vocabulary reduction and release our implementation for reproducibility.", "date": "2025-12-27", "categories": []} {"id": "2507.04716", "title": "Optimal Model Selection for Conformalized Robust Optimization", "abstract": "In decision-making under uncertainty, Contextual Robust Optimization (CRO) provides reliability by minimizing the worst-case decision loss over a prediction set. While recent advances use conformal prediction to construct prediction sets for machine learning models, the downstream decisions critically depend on model selection. This paper introduces novel model selection frameworks for CRO that unify robustness control with decision risk minimization. We first propose Conformalized Robust Optimization with Model Selection (CROMS), a framework that selects the model to approximately minimize the averaged decision risk in CRO solutions. Given the target robustness level 1-\\alpha, we present a computationally efficient algorithm called E-CROMS, which achieves asymptotic robustness control and decision optimality. To correct the control bias in finite samples, we further develop two algorithms: F-CROMS, which ensures a 1-\\alpha robustness but requires searching the label space; and J-CROMS, which offers lower computational cost while achieving a 1-2\\alpha robustness. Furthermore, we extend the CROMS framework to the individualized setting, where model selection is performed by minimizing the conditional decision risk given the covariates of the test data. This framework advances conformal prediction methodology by enabling covariate-aware model selection. Numerical results demonstrate significant improvements in decision efficiency across diverse synthetic and real-world applications, outperforming baseline approaches.", "date": "2025-07-07", "categories": []} {"id": "2511.18562", "title": "Ensuring Calibration Robustness in Split Conformal Prediction Under Adversarial Attacks", "abstract": "Conformal prediction (CP) provides distribution-free, finite-sample coverage guarantees but critically relies on exchangeability, a condition often violated under distribution shift. We study the robustness of split conformal prediction under adversarial perturbations at test time, focusing on both coverage validity and the resulting prediction set size. Our theoretical analysis characterizes how the strength of adversarial perturbations during calibration affects coverage guarantees under adversarial test conditions. We further examine the impact of adversarial training at the model-training stage. Extensive experiments support our theory: (i) Prediction coverage varies monotonically with the calibration-time attack strength, enabling the use of nonzero calibration-time attack to predictably control coverage under adversarial tests; (ii) target coverage can hold over a range of test-time attacks: with a suitable calibration attack, coverage stays within any chosen tolerance band across a contiguous set of perturbation levels; and (iii) adversarial training at the training stage produces tighter prediction sets that retain high informativeness.", "date": "2025-11-23", "categories": []} {"id": "2512.17048", "title": "Another Fit Bites the Dust: Conformal Prediction as a Calibration Standard for Machine Learning in High-Energy Physics", "abstract": "Machine-learning techniques are essential in modern collider research, yet their probabilistic outputs often lack calibrated uncertainty estimates and finite-sample guarantees, limiting their direct use in statistical inference and decision-making. Conformal prediction (CP) provides a simple, distribution-free framework for calibrating arbitrary predictive models without retraining, yielding rigorous uncertainty quantification with finite-sample coverage guarantees under minimal exchangeability assumptions, without reliance on asymptotics, limit theorems, or Gaussian approximations. In this work, we investigate CP as a unifying calibration layer for machine-learning applications in high-energy physics. Using publicly available collider datasets and a diverse set of models, we show that a single conformal formalism can be applied across regression, binary and multi-class classification, anomaly detection, and generative modelling, converting raw model outputs into statistically valid prediction sets, typicality regions, and p-values with controlled false-positive rates. While conformal prediction does not improve raw model performance, it enforces honest uncertainty quantification and transparent error control. We argue that conformal calibration should be adopted as a standard component of machine-learning pipelines in collider physics, enabling reliable interpretation, robust comparisons, and principled statistical decisions in experimental and phenomenological analyses.", "date": "2025-12-18", "categories": []} {"id": "2512.14727", "title": "A Critical Perspective on Finite Sample Conformal Prediction Theory in Medical Applications", "abstract": "Machine learning (ML) is transforming healthcare, but safe clinical decisions demand reliable uncertainty estimates that standard ML models fail to provide. Conformal prediction (CP) is a popular tool that allows users to turn heuristic uncertainty estimates into uncertainty estimates with statistical guarantees. CP works by converting predictions of a ML model, together with a calibration sample, into prediction sets that are guaranteed to contain the true label with any desired probability. An often cited advantage is that CP theory holds for calibration samples of arbitrary size, suggesting that uncertainty estimates with practically meaningful statistical guarantees can be achieved even if only small calibration sets are available. We question this promise by showing that, although the statistical guarantees hold for calibration sets of arbitrary size, the practical utility of these guarantees does highly depend on the size of the calibration set. This observation is relevant in medical domains because data is often scarce and obtaining large calibration sets is therefore infeasible. We corroborate our critique in an empirical demonstration on a medical image classification task.", "date": "2025-12-09", "categories": []} {"id": "2512.04566", "title": "Reliable Statistical Guarantees for Conformal Predictors with Small Datasets", "abstract": "Surrogate models (including deep neural networks and other machine learning algorithms in supervised learning) are capable of approximating arbitrarily complex, high-dimensional input-output problems in science and engineering, but require a thorough data-agnostic uncertainty quantification analysis before these can be deployed for any safety-critical application. The standard approach for data-agnostic uncertainty quantification is to use conformal prediction (CP), a well-established framework to build uncertainty models with proven statistical guarantees that do not assume any shape for the error distribution of the surrogate model. However, since the classic statistical guarantee offered by CP is given in terms of bounds for the marginal coverage, for small calibration set sizes (which are frequent in realistic surrogate modelling that aims to quantify error at different regions), the potentially strong dispersion of the coverage distribution around its average negatively impacts the relevance of the uncertainty model's statistical guarantee, often obtaining coverages below the expected value, resulting in a less applicable framework. After providing a gentle presentation of uncertainty quantification for surrogate models for machine learning practitioners, in this paper we bridge the gap by proposing a new statistical guarantee that offers probabilistic information for the coverage of a single conformal predictor. We show that the proposed framework converges to the standard solution offered by CP for large calibration set sizes and, unlike the classic guarantee, still offers relevant information about the coverage of a conformal predictor for small data sizes. We validate the methodology in a suite of examples, and implement an open access software solution that can be used alongside common conformal prediction libraries to obtain uncertainty models that fulfil the new guarantee.", "date": "2025-12-04", "categories": []} {"id": "2512.14221", "title": "Weighted Conformal Prediction Provides Adaptive and Valid Mask-Conditional Coverage for General Missing Data Mechanisms", "abstract": "Conformal prediction (CP) offers a principled framework for uncertainty quantification, but it fails to guarantee coverage when faced with missing covariates. In addressing the heterogeneity induced by various missing patterns, Mask-Conditional Valid (MCV) Coverage has emerged as a more desirable property than Marginal Coverage. In this work, we adapt split CP to handle missing values by proposing a preimpute-mask-then-correct framework that can offer valid coverage. We show that our method provides guaranteed Marginal Coverage and Mask-Conditional Validity for general missing data mechanisms. A key component of our approach is a reweighted conformal prediction procedure that corrects the prediction sets after distributional imputation (multiple imputation) of the calibration dataset, making our method compatible with standard imputation pipelines. We derive two algorithms, and we show that they are approximately marginally valid and MCV. We evaluate them on synthetic and real-world datasets. It reduces significantly the width of prediction intervals w.r.t standard MCV methods, while maintaining the target guarantees.", "date": "2025-12-16", "categories": []} {"id": "2506.16553", "title": "One Sample is Enough to Make Conformal Prediction Robust", "abstract": "For any black-box model, conformal prediction (CP) returns prediction sets guaranteed to include the true label with high adjustable probability. Robust CP (RCP) extends the guarantee to the worst case noise up to a pre-defined magnitude. For RCP, a well-established approach is to use randomized smoothing since it is applicable to any black-box model and provides smaller sets compared to deterministic methods. However, smoothing-based robustness requires many model forward passes per each input which is computationally expensive. We show that conformal prediction attains some robustness even with a single forward pass on a randomly perturbed input. Using any binary certificate we propose a single sample robust CP (RCP1). Our approach returns robust sets with smaller average set size compared to SOTA methods which use many (e.g. 100) passes per input. Our key insight is to certify the conformal procedure itself rather than individual conformity scores. Our approach is agnostic to the task (classification and regression). We further extend our approach to smoothing-based robust conformal risk control.", "date": "2025-06-19", "categories": []} {"id": "2505.15721", "title": "Privacy-Preserving Conformal Prediction Under Local Differential Privacy", "abstract": "Conformal prediction (CP) provides sets of candidate classes with a guaranteed probability of containing the true class. However, it typically relies on a calibration set with clean labels. We address privacy-sensitive scenarios where the aggregator is untrusted and can only access a perturbed version of the true labels. We propose two complementary approaches under local differential privacy (LDP). In the first approach, users do not access the model but instead provide their input features and a perturbed label using a k-ary randomized response. In the second approach, which enforces stricter privacy constraints, users add noise to their conformity score by binary search response. This method requires access to the classification model but preserves both data and label privacy. Both approaches compute the conformal threshold directly from noisy data without accessing the true labels. We prove finite-sample coverage guarantees and demonstrate robust coverage even under severe randomization. This approach unifies strong local privacy with predictive uncertainty control, making it well-suited for sensitive applications such as medical imaging or large language model queries, regardless of whether users can (or are willing to) compute their own scores.", "date": "2025-05-21", "categories": []} {"id": "2512.05611", "title": "Design-marginal calibration of Gaussian process predictive distributions: Bayesian and conformal approaches", "abstract": "We study the calibration of Gaussian process (GP) predictive distributions in the interpolation setting from a design-marginal perspective. Conditioning on the data and averaging over a design measure \\mu, we formalize \\mu-coverage for central intervals and \\mu-probabilistic calibration through randomized probability integral transforms. We introduce two methods. cps-gp adapts conformal predictive systems to GP interpolation using standardized leave-one-out residuals, yielding stepwise predictive distributions with finite-sample marginal calibration. bcr-gp retains the GP posterior mean and replaces the Gaussian residual by a generalized normal model fitted to cross-validated standardized residuals. A Bayesian selection rule-based either on a posterior upper quantile of the variance for conservative prediction or on a cross-posterior Kolmogorov-Smirnov criterion for probabilistic calibration-controls dispersion and tail behavior while producing smooth predictive distributions suitable for sequential design. Numerical experiments on benchmark functions compare cps-gp, bcr-gp, Jackknife+ for GPs, and the full conformal Gaussian process, using calibration metrics (coverage, Kolmogorov-Smirnov, integral absolute error) and accuracy or sharpness through the scaled continuous ranked probability score.", "date": "2025-12-05", "categories": []} {"id": "2501.11384", "title": "Transductive Conformal Inference for Full Ranking", "abstract": "We introduce a method based on Conformal Prediction (CP) to quantify the uncertainty of full ranking algorithms. We focus on a specific scenario where $n+m$ items are to be ranked by some ``black box'' algorithm. It is assumed that the relative (ground truth) ranking of $n$ of them is known. The objective is then to quantify the error made by the algorithm on the ranks of the $m$ new items among the total $(n+m)$. In such a setting, the true ranks of the $n$ original items in the total $(n+m)$ depend on the (unknown) true ranks of the $m$ new ones. Consequently, we have no direct access to a calibration set to apply a classical CP method. To address this challenge, we propose to construct distribution-free bounds of the unknown conformity scores using recent results on the distribution of conformal p-values. Using these scores upper bounds, we provide valid prediction sets for the rank of any item. We also control the false coverage proportion, a crucial quantity when dealing with multiple prediction sets. Finally, we empirically show on both synthetic and real data the efficiency of our CP method for state-of-the-art algorithms such as RankNet or LambdaMart.", "date": "2025-01-20", "categories": []} {"id": "2507.07150", "title": "Class conditional conformal prediction for multiple inputs by p-value aggregation", "abstract": "Conformal prediction methods are statistical tools designed to quantify uncertainty and generate predictive sets with guaranteed coverage probabilities. This work introduces an innovative refinement to these methods for classification tasks, specifically tailored for scenarios where multiple observations (multi-inputs) of a single instance are available at prediction time. Our approach is particularly motivated by applications in citizen science, where multiple images of the same plant or animal are captured by individuals. Our method integrates the information from each observation into conformal prediction, enabling a reduction in the size of the predicted label set while preserving the required class-conditional coverage guarantee. The approach is based on the aggregation of conformal p-values computed from each observation of a multi-input. By exploiting the exact distribution of these p-values, we propose a general aggregation framework using an abstract scoring function, encompassing many classical statistical tools. Knowledge of this distribution also enables refined versions of standard strategies, such as majority voting. We evaluate our method on simulated and real data, with a particular focus on Pl@ntNet, a prominent citizen science platform that facilitates the collection and identification of plant species through user-submitted images.", "date": "2025-07-09", "categories": []} {"id": "2512.04016", "title": "TARA Test-by-Adaptive-Ranks for Quantum Anomaly Detection with Conformal Prediction Guarantees", "abstract": "Quantum key distribution (QKD) security fundamentally relies on the ability to distinguish genuine quantum correlations from classical eavesdropper simulations, yet existing certification methods lack rigorous statistical guarantees under finite-sample conditions and adversarial scenarios. We introduce TARA (Test by Adaptive Ranks), a novel framework combining conformal prediction with sequential martingale testing for quantum anomaly detection that provides distribution-free validity guarantees. TARA offers two complementary approaches. TARA k, based on Kolmogorov Smirnov calibration against local hidden variable (LHV) null distributions, achieving ROC AUC = 0.96 for quantum-classical discrimination. And TARA-m, employing betting martingales for streaming detection with anytime valid type I error control that enables real time monitoring of quantum channels. We establish theoretical guarantees proving that under (context conditional) exchangeability, conformal p-values remain uniformly distributed even for strongly contextual quantum data, confirming that quantum contextuality does not break conformal prediction validity a result with implications beyond quantum certification to any application of distribution-free methods to nonclassical data. Extensive validation on both IBM Torino (superconducting, CHSH = 2.725) and IonQ Forte Enterprise (trapped ion, CHSH = 2.716) quantum processors demonstrates cross-platform robustness, achieving 36% security margins above the classical CHSH bound of 2. Critically, our framework reveals a methodological concern affecting quantum certification more broadly: same-distribution calibration can inflate detection performance by up to 44 percentage points compared to proper cross-distribution calibration, suggesting that prior quantum certification studies using standard train test splits may have systematically overestimated adversarial robustness.", "date": "2025-12-03", "categories": []} {"id": "2512.02704", "title": "Conformal Correction for Efficiency May be at Odds with Entropy", "abstract": "Conformal prediction (CP) provides a comprehensive framework to produce statistically rigorous uncertainty sets for black-box machine learning models. To further improve the efficiency of CP, conformal correction is proposed to fine-tune or wrap the base model with an extra module using a conformal-aware inefficiency loss. In this work, we empirically and theoretically identify a trade-off between the CP efficiency and the entropy of model prediction. We then propose an entropy-constrained conformal correction method, exploring a better Pareto optimum between efficiency and entropy. Extensive experimental results on both computer vision and graph datasets demonstrate the efficacy of the proposed method. For instance, it can significantly improve the efficiency of state-of-the-art CP methods by up to 34.4%, given an entropy threshold.", "date": "2025-12-02", "categories": []} {"id": "2403.07728", "title": "CAP: A General Algorithm for Online Selective Conformal Prediction with FCR Control", "abstract": "We study the problem of post-selection predictive inference in an online fashion. To avoid devoting resources to unimportant units, a preliminary selection of the current individual before reporting its prediction interval is common and meaningful in online predictive tasks. Since the online selection causes a temporal multiplicity in the selected prediction intervals, it is important to control the real-time false coverage-statement rate (FCR) which measures the overall miscoverage level. We develop a general framework named CAP (Calibration after Adaptive Pick) that performs an adaptive pick rule on historical data to construct a calibration set if the current individual is selected and then outputs a conformal prediction interval for the unobserved label. We provide tractable procedures for constructing the calibration set for popular online selection rules. We proved that CAP can achieve an exact selection-conditional coverage guarantee in the finite-sample and distribution-free regimes. To account for the distribution shift in online data, we also embed CAP into some recent dynamic conformal prediction algorithms and show that the proposed method can deliver long-run FCR control. Numerical results on both synthetic and real data corroborate that CAP can effectively control FCR around the target level and yield more narrowed prediction intervals over existing baselines across various settings.", "date": "2024-03-12", "categories": []} {"id": "2411.19523", "title": "Localized Conformal Multi-Quantile Regression", "abstract": "Standard conformal prediction methods guarantee marginal coverage but often produce inefficient intervals that fail to adapt to local heteroscedasticity, while recent localized approaches often struggle to maintain validity across distinct subpopulations with varying noise profiles. To address these challenges, we introduce Localized Conformal Multi-Quantile Regression (LCMQR), a novel framework that synergizes multi-quantile information with kernel-based localization to construct efficient and adaptive prediction intervals. Theoretically, we resolve an inconsistency in Conformalized Composite Quantile Regression (CCQR) by proving that our consistent Average-then-Max scoring mechanism systematically yields tighter intervals than the Max-then-Average approach used in prior work. For heterogeneous environments, we extend this framework to Group-Calibrated LCMQR (GC-LCMQR) via a stratified calibration step that guarantees finite-sample validity within distinct subgroups. Experiments on benchmark datasets and an Individual Treatment Effect (ITE) task demonstrate that LCMQR achieves superior efficiency on standard benchmarks, while GC-LCMQR uniquely achieves group-level coverage for target subgroups in mixture populations where baselines fail.", "date": "2024-11-29", "categories": []} {"id": "2509.02844", "title": "Conformal Prediction for Time-series Forecasting with Change Points", "abstract": "Conformal prediction has been explored as a general and efficient way to provide uncertainty quantification for time series. However, current methods struggle to handle time series data with change points - sudden shifts in the underlying data-generating process. In this paper, we propose a novel Conformal Prediction for Time-series with Change points (CPTC) algorithm, addressing this gap by integrating a model to predict the underlying state with online conformal prediction to model uncertainties in non-stationary time series. We prove CPTC's validity and improved adaptivity in the time series setting under minimum assumptions, and demonstrate CPTC's practical effectiveness on 6 synthetic and real-world datasets, showing improved validity and adaptivity compared to state-of-the-art baselines.", "date": "2025-09-02", "categories": []} {"id": "2407.07700", "title": "Split Conformal Prediction under Data Contamination", "abstract": "Conformal prediction is a non-parametric technique for constructing prediction intervals or sets from arbitrary predictive models under the assumption that the data is exchangeable. It is popular as it comes with theoretical guarantees on the marginal coverage of the prediction sets and the split conformal prediction variant has a very low computational cost compared to model training. We study the robustness of split conformal prediction in a data contamination setting, where we assume a small fraction of the calibration scores are drawn from a different distribution than the bulk. We quantify the impact of the corrupted data on the coverage and efficiency of the constructed sets when evaluated on \"clean\" test points, and verify our results with numerical experiments. Moreover, we propose an adjustment in the classification setting which we call Contamination Robust Conformal Prediction, and verify the efficacy of our approach using both synthetic and real datasets.", "date": "2024-07-10", "categories": []} {"id": "2511.20811", "title": "Conformal Safety Monitoring for Flight Testing: A Case Study in Data-Driven Safety Learning", "abstract": "We develop a data-driven approach for runtime safety monitoring in flight testing, where pilots perform maneuvers on aircraft with uncertain parameters. Because safety violations can arise unexpectedly as a result of these uncertainties, pilots need clear, preemptive criteria to abort the maneuver in advance of safety violation. To solve this problem, we use offline stochastic trajectory simulation to learn a calibrated statistical model of the short-term safety risk facing pilots. We use flight testing as a motivating example for data-driven learning/monitoring of safety due to its inherent safety risk, uncertainty, and human-interaction. However, our approach consists of three broadly-applicable components: a model to predict future state from recent observations, a nearest neighbor model to classify the safety of the predicted state, and classifier calibration via conformal prediction. We evaluate our method on a flight dynamics model with uncertain parameters, demonstrating its ability to reliably identify unsafe scenarios, match theoretical guarantees, and outperform baseline approaches in preemptive classification of risk.", "date": "2025-11-25", "categories": []} {"id": "2511.13533", "title": "Minimax Multi-Target Conformal Prediction with Applications to Imaging Inverse Problems", "abstract": "In ill-posed imaging inverse problems, uncertainty quantification remains a fundamental challenge, especially in safety-critical applications. Recently, conformal prediction has been used to quantify the uncertainty that the inverse problem contributes to downstream tasks like image classification, image quality assessment, fat mass quantification, etc. While existing works handle only a scalar estimation target, practical applications often involve multiple targets. In response, we propose an asymptotically minimax approach to multi-target conformal prediction that provides tight prediction intervals while ensuring joint marginal coverage. We then outline how our minimax approach can be applied to multi-metric blind image quality assessment, multi-task uncertainty quantification, and multi-round measurement acquisition. Finally, we numerically demonstrate the benefits of our minimax method, relative to existing multi-target conformal prediction methods, using both synthetic and magnetic resonance imaging (MRI) data. Code is available at https://github.com/jwen307/multi_target_minimax.", "date": "2025-11-17", "categories": []} {"id": "2511.17861", "title": "Cost-Sensitive Conformal Training with Provably Controllable Learning Bounds", "abstract": "Conformal prediction (CP) is a general framework to quantify the predictive uncertainty of machine learning models that uses a set prediction to include the true label with a valid probability. To align the uncertainty measured by CP, conformal training methods minimize the size of the prediction sets. A typical way is to use a surrogate indicator function, usually Sigmoid or Gaussian error function. However, these surrogate functions do not have a uniform error bound to the indicator function, leading to uncontrollable learning bounds. In this paper, we propose a simple cost-sensitive conformal training algorithm that does not rely on the indicator approximation mechanism. Specifically, we theoretically show that minimizing the expected size of prediction sets is upper bounded by the expected rank of true labels. To this end, we develop a rank weighting strategy that assigns the weight using the rank of true label on each data sample. Our analysis provably demonstrates the tightness between the proposed weighted objective and the expected size of conformal prediction sets. Extensive experiments verify the validity of our theoretical insights, and superior empirical performance over other conformal training in terms of predictive efficiency with 21.38% reduction for average prediction set size.", "date": "2025-11-22", "categories": []} {"id": "2509.05826", "title": "Performance of Conformal Prediction in Capturing Aleatoric Uncertainty", "abstract": "Conformal prediction is a model-agnostic approach to generating prediction sets that cover the true class with a high probability. Although its prediction set size is expected to capture aleatoric uncertainty, there is a lack of evidence regarding its effectiveness. The literature presents that prediction set size can upper-bound aleatoric uncertainty or that prediction sets are larger for difficult instances and smaller for easy ones, but a validation of this attribute of conformal predictors is missing. This work investigates how effectively conformal predictors quantify aleatoric uncertainty, specifically the inherent ambiguity in datasets caused by overlapping classes. We perform this by measuring the correlation between prediction set sizes and the number of distinct labels assigned by human annotators per instance. We further assess the similarity between prediction sets and human-provided annotations. We use three conformal prediction approaches to generate prediction sets for eight deep learning models trained on four datasets. The datasets contain annotations from multiple human annotators (ranging from five to fifty participants) per instance, enabling the identification of class overlap. We show that the vast majority of the conformal prediction outputs show a very weak to weak correlation with human annotations, with only a few showing moderate correlation. These findings underscore the necessity of critically reassessing the prediction sets generated using conformal predictors. While they can provide a higher coverage of the true classes, their capability in capturing aleatoric uncertainty and generating sets that align with human annotations remains limited.", "date": "2025-09-06", "categories": []} {"id": "2511.16845", "title": "Provably Minimum-Length Conformal Prediction Sets for Ordinal Classification", "abstract": "Ordinal classification has been widely applied in many high-stakes applications, e.g., medical imaging and diagnosis, where reliable uncertainty quantification (UQ) is essential for decision making. Conformal prediction (CP) is a general UQ framework that provides statistically valid guarantees, which is especially useful in practice. However, prior ordinal CP methods mainly focus on heuristic algorithms or restrictively require the underlying model to predict a unimodal distribution over ordinal labels. Consequently, they provide limited insight into coverage-efficiency trade-offs, or a model-agnostic and distribution-free nature favored by CP methods. To this end, we fill this gap by propose an ordinal-CP method that is model-agnostic and provides instance-level optimal prediction intervals. Specifically, we formulate conformal ordinal classification as a minimum-length covering problem at the instance level. To solve this problem, we develop a sliding-window algorithm that is optimal on each calibration data, with only a linear time complexity in K, the number of label candidates. The local optimality per instance further also improves predictive efficiency in expectation. Moreover, we propose a length-regularized variant that shrinks prediction set size while preserving coverage. Experiments on four benchmark datasets from diverse domains are conducted to demonstrate the significantly improved predictive efficiency of the proposed methods over baselines (by 15% decrease on average over four datasets).", "date": "2025-11-20", "categories": []} {"id": "2511.07899", "title": "Statistically Assuring Safety of Control Systems using Ensembles of Safety Filters and Conformal Prediction", "abstract": "Safety assurance is a fundamental requirement for deploying learning-enabled autonomous systems. Hamilton-Jacobi (HJ) reachability analysis is a fundamental method for formally verifying safety and generating safe controllers. However, computing the HJ value function that characterizes the backward reachable set (BRS) of a set of user-defined failure states is computationally expensive, especially for high-dimensional systems, motivating the use of reinforcement learning approaches to approximate the value function. Unfortunately, a learned value function and its corresponding safe policy are not guaranteed to be correct. The learned value function evaluated at a given state may not be equal to the actual safety return achieved by following the learned safe policy. To address this challenge, we introduce a conformal prediction-based (CP) framework that bounds such uncertainty. We leverage CP to provide probabilistic safety guarantees when using learned HJ value functions and policies to prevent control systems from reaching failure states. Specifically, we use CP to calibrate the switching between the unsafe nominal controller and the learned HJ-based safe policy and to derive safety guarantees under this switched policy. We also investigate using an ensemble of independently trained HJ value functions as a safety filter and compare this ensemble approach to using individual value functions alone.", "date": "2025-11-11", "categories": []} {"id": "2511.15838", "title": "Attention-Based Feature Online Conformal Prediction for Time Series", "abstract": "Online conformal prediction (OCP) wraps around any pre-trained predictor to produce prediction sets with coverage guarantees that hold irrespective of temporal dependencies or distribution shifts. However, standard OCP faces two key limitations: it operates in the output space using simple nonconformity (NC) scores, and it treats all historical observations uniformly when estimating quantiles. This paper introduces attention-based feature OCP (AFOCP), which addresses both limitations through two key innovations. First, AFOCP operates in the feature space of pre-trained neural networks, leveraging learned representations to construct more compact prediction sets by concentrating on task-relevant information while suppressing nuisance variation. Second, AFOCP incorporates an attention mechanism that adaptively weights historical observations based on their relevance to the current test point, effectively handling non-stationarity and distribution shifts. We provide theoretical guarantees showing that AFOCP maintains long-term coverage while provably achieving smaller prediction intervals than standard OCP under mild regularity conditions. Extensive experiments on synthetic and real-world time series datasets demonstrate that AFOCP consistently reduces the size of prediction intervals by as much as $88\\%$ as compared to OCP, while maintaining target coverage levels, validating the benefits of both feature-space calibration and attention-based adaptive weighting.", "date": "2025-11-19", "categories": []} {"id": "2509.01840", "title": "Optimizing In-Context Learning for Efficient Full Conformal Prediction", "abstract": "Reliable uncertainty quantification is critical for trustworthy AI. Conformal Prediction (CP) provides prediction sets with distribution-free coverage guarantees, but its two main variants face complementary limitations. Split CP (SCP) suffers from data inefficiency due to dataset partitioning, while full CP (FCP) improves data efficiency at the cost of prohibitive retraining complexity. Recent approaches based on meta-learning or in-context learning (ICL) partially mitigate these drawbacks. However, they rely on training procedures not specifically tailored to CP, which may yield large prediction sets. We introduce an efficient FCP framework, termed enhanced ICL-based FCP (E-ICL+FCP), which employs a permutation-invariant Transformer-based ICL model trained with a CP-aware loss. By simulating the multiple retrained models required by FCP without actual retraining, E-ICL+FCP preserves coverage while markedly reducing both inefficiency and computational overhead. Experiments on synthetic and real tasks demonstrate that E-ICL+FCP attains superior efficiency-coverage trade-offs compared to existing SCP and FCP baselines.", "date": "2025-09-01", "categories": []} {"id": "2511.15146", "title": "Beyond Uncertainty Sets: Leveraging Optimal Transport to Extend Conformal Predictive Distribution to Multivariate Settings", "abstract": "Conformal prediction (CP) constructs uncertainty sets for model outputs with finite-sample coverage guarantees. A candidate output is included in the prediction set if its non-conformity score is not considered extreme relative to the scores observed on a set of calibration examples. However, this procedure is only straightforward when scores are scalar-valued, which has limited CP to real-valued scores or ad-hoc reductions to one dimension. The problem of ordering vectors has been studied via optimal transport (OT), which provides a principled method for defining vector-ranks and multivariate quantile regions, though typically with only asymptotic coverage guarantees. We restore finite-sample, distribution-free coverage by conformalizing the vector-valued OT quantile region. Here, a candidate's rank is defined via a transport map computed for the calibration scores augmented with that candidate's score. This defines a continuum of OT problems for which we prove that the resulting optimal assignment is piecewise-constant across a fixed polyhedral partition of the score space. This allows us to characterize the entire prediction set tractably, and provides the machinery to address a deeper limitation of prediction sets: that they only indicate which outcomes are plausible, but not their relative likelihood. In one dimension, conformal predictive distributions (CPDs) fill this gap by producing a predictive distribution with finite-sample calibration. Extending CPDs beyond one dimension remained an open problem. We construct, to our knowledge, the first multivariate CPDs with finite-sample calibration, i.e., they define a valid multivariate distribution where any derived uncertainty region automatically has guaranteed coverage. We present both conservative and exact randomized versions, the latter resulting in a multivariate generalization of the classical Dempster-Hill procedure.", "date": "2025-11-19", "categories": []} {"id": "2511.13911", "title": "Uncertainty-Calibrated Prediction of Randomly-Timed Biomarker Trajectories with Conformal Bands", "abstract": "Despite recent progress in predicting biomarker trajectories from real clinical data, uncertainty in the predictions poses high-stakes risks (e.g., misdiagnosis) that limit their clinical deployment. To enable safe and reliable use of such predictions in healthcare, we introduce a conformal method for uncertainty-calibrated prediction of biomarker trajectories resulting from randomly-timed clinical visits of patients. Our approach extends conformal prediction to the setting of randomly-timed trajectories via a novel nonconformity score that produces prediction bands guaranteed to cover the unknown biomarker trajectories with a user-prescribed probability. We apply our method across a wide range of standard and state-of-the-art predictors for two well-established brain biomarkers of Alzheimer's disease, using neuroimaging data from real clinical studies. We observe that our conformal prediction bands consistently achieve the desired coverage, while also being tighter than baseline prediction bands. To further account for population heterogeneity, we develop group-conditional conformal bands and test their coverage guarantees across various demographic and clinically relevant subpopulations. Moreover, we demonstrate the clinical utility of our conformal bands in identifying subjects at high risk of progression to Alzheimer's disease. Specifically, we introduce an uncertainty-calibrated risk score that enables the identification of 17.5% more high-risk subjects compared to standard risk scores, highlighting the value of uncertainty calibration in real-world clinical decision making. Our code is available at github.com/vatass/ConformalBiomarkerTrajectories.", "date": "2025-11-17", "categories": []} {"id": "2511.04534", "title": "Uncertainty Quantification for Reduced-Order Surrogate Models Applied to Cloud Microphysics", "abstract": "Reduced-order models (ROMs) can efficiently simulate high-dimensional physical systems but lack robust uncertainty quantification methods. Existing approaches are frequently architecture- or training-specific, which limits flexibility and generalization. We introduce a post hoc, model-agnostic framework for predictive uncertainty quantification in latent space ROMs that requires no modification to the underlying architecture or training procedure. Using conformal prediction, our approach estimates statistical prediction intervals for multiple components of the ROM pipeline: latent dynamics, reconstruction, and end-to-end predictions. We demonstrate the method on a latent space dynamical model for cloud microphysics, where it accurately predicts the evolution of droplet-size distributions and quantifies uncertainty across the ROM pipeline.", "date": "2025-11-06", "categories": []} {"id": "2511.11472", "title": "Quantifying and Improving Adaptivity in Conformal Prediction through Input Transformations", "abstract": "Conformal prediction constructs a set of labels instead of a single point prediction, while providing a probabilistic coverage guarantee. Beyond the coverage guarantee, adaptiveness to example difficulty is an important property. It means that the method should produce larger prediction sets for more difficult examples, and smaller ones for easier examples. Existing evaluation methods for adaptiveness typically analyze coverage rate violation or average set size across bins of examples grouped by difficulty. However, these approaches often suffer from imbalanced binning, which can lead to inaccurate estimates of coverage or set size. To address this issue, we propose a binning method that leverages input transformations to sort examples by difficulty, followed by uniform-mass binning. Building on this binning, we introduce two metrics to better evaluate adaptiveness. These metrics provide more reliable estimates of coverage rate violation and average set size due to balanced binning, leading to more accurate adaptivity assessment. Through experiments, we demonstrate that our proposed metric correlates more strongly with the desired adaptiveness property compared to existing ones. Furthermore, motivated by our findings, we propose a new adaptive prediction set algorithm that groups examples by estimated difficulty and applies group-conditional conformal prediction. This allows us to determine appropriate thresholds for each group. Experimental results on both (a) an Image Classification (ImageNet) (b) a medical task (visual acuity prediction) show that our method outperforms existing approaches according to the new metrics.", "date": "2025-11-14", "categories": []} {"id": "2511.12065", "title": "Aggregating Conformal Prediction Sets via {\\alpha}-Allocation", "abstract": "Conformal prediction offers a distribution-free framework for constructing prediction sets with finite-sample coverage. Yet, efficiently leveraging multiple conformity scores to reduce prediction set size remains a major open challenge. Instead of selecting a single best score, this work introduces a principled aggregation strategy, COnfidence-Level Allocation (COLA), that optimally allocates confidence levels across multiple conformal prediction sets to minimize empirical set size while maintaining provable coverage. Two variants are further developed, COLA-s and COLA-f, which guarantee finite-sample marginal coverage via sample splitting and full conformalization, respectively. In addition, we develop COLA-l, an individualized allocation strategy that promotes local size efficiency while achieving asymptotic conditional coverage. Extensive experiments on synthetic and real-world datasets demonstrate that COLA achieves considerably smaller prediction sets than state-of-the-art baselines while maintaining valid coverage.", "date": "2025-11-15", "categories": []} {"id": "2511.13608", "title": "A Gentle Introduction to Conformal Time Series Forecasting", "abstract": "Conformal prediction is a powerful post-hoc framework for uncertainty quantification that provides distribution-free coverage guarantees. However, these guarantees crucially rely on the assumption of exchangeability. This assumption is fundamentally violated in time series data, where temporal dependence and distributional shifts are pervasive. As a result, classical split-conformal methods may yield prediction intervals that fail to maintain nominal validity. This review unifies recent advances in conformal forecasting methods specifically designed to address nonexchangeable data. We first present a theoretical foundation, deriving finite-sample guarantees for split-conformal prediction under mild weak-dependence conditions. We then survey and classify state-of-the-art approaches that mitigate serial dependence by reweighting calibration data, dynamically updating residual distributions, or adaptively tuning target coverage levels in real time. Finally, we present a comprehensive simulation study that compares these techniques in terms of empirical coverage, interval width, and computational cost, highlighting practical trade-offs and open research directions.", "date": "2025-11-17", "categories": []} {"id": "2410.16493", "title": "Building Conformal Prediction Intervals with Approximate Message Passing", "abstract": "Conformal prediction has emerged as a powerful tool for building prediction intervals that are valid in a distribution-free way. However, its evaluation may be computationally costly, especially in the high-dimensional setting where the dimensionality and sample sizes are both large and of comparable magnitudes. To address this challenge in the context of generalized linear regression, we propose a novel algorithm based on Approximate Message Passing (AMP) to accelerate the computation of prediction intervals using full conformal prediction, by approximating the computation of conformity scores. Our work bridges a gap between modern uncertainty quantification techniques and tools for high-dimensional problems involving the AMP algorithm. We evaluate our method on both synthetic and real data, and show that it produces prediction intervals that are close to the baseline methods, while being orders of magnitude faster. Additionally, in the high-dimensional limit and under assumptions on the data distribution, the conformity scores computed by AMP converge to the one computed exactly, which allows theoretical study and benchmarking of conformal methods in high dimensions.", "date": "2024-10-21", "categories": []} {"id": "2511.08808", "title": "Effects of label noise on the classification of outlier observations", "abstract": "This study investigates the impact of adding noise to the training set classes in classification tasks using the BCOPS algorithm (Balanced and Conformal Optimized Prediction Sets), proposed by Guan & Tibshirani (2022). The BCOPS algorithm is an application of conformal prediction combined with a machine learning method to construct prediction sets such that the probability of the true class being included in the prediction set for a test observation meets a specified coverage guarantee. An observation is considered an outlier if its true class is not present in the training set. The study employs both synthetic and real datasets and conducts experiments to evaluate the prediction abstention rate for outlier observations and the model's robustness in this previously untested scenario. The results indicate that the addition of noise, even in small amounts, can have a significant effect on model performance.", "date": "2025-11-11", "categories": []} {"id": "2503.04981", "title": "Topology-Aware Conformal Prediction for Stream Networks", "abstract": "Stream networks, a unique class of spatiotemporal graphs, exhibit complex directional flow constraints and evolving dependencies, making uncertainty quantification a critical yet challenging task. Traditional conformal prediction methods struggle in this setting due to the need for joint predictions across multiple interdependent locations and the intricate spatio-temporal dependencies inherent in stream networks. Existing approaches either neglect dependencies, leading to overly conservative predictions, or rely solely on data-driven estimations, failing to capture the rich topological structure of the network. To address these challenges, we propose Spatio-Temporal Adaptive Conformal Inference (\\texttt{STACI}), a novel framework that integrates network topology and temporal dynamics into the conformal prediction framework. \\texttt{STACI} introduces a topology-aware nonconformity score that respects directional flow constraints and dynamically adjusts prediction sets to account for temporal distributional shifts. We provide theoretical guarantees on the validity of our approach and demonstrate its superior performance on both synthetic and real-world datasets. Our results show that \\texttt{STACI} effectively balances prediction efficiency and coverage, outperforming existing conformal prediction methods for stream networks.", "date": "2025-03-06", "categories": []} {"id": "2506.05583", "title": "Conformal Prediction Adaptive to Unknown Subpopulation Shifts", "abstract": "Conformal prediction is widely used to equip black-box machine learning models with uncertainty quantification, offering formal coverage guarantees under exchangeable data. However, these guarantees fail when faced with subpopulation shifts, where the test environment contains a different mix of subpopulations than the calibration data. In this work, we focus on unknown subpopulation shifts where we are not given group-information i.e. the subpopulation labels of datapoints have to be inferred. We propose new methods that provably adapt conformal prediction to such shifts, ensuring valid coverage without explicit knowledge of subpopulation structure. While existing methods in similar setups assume perfect subpopulation labels, our framework explicitly relaxes this requirement and characterizes conditions where formal coverage guarantees remain feasible. Further, our algorithms scale to high-dimensional settings and remain practical in realistic machine learning tasks. Extensive experiments on vision (with vision transformers) and language (with large language models) benchmarks demonstrate that our methods reliably maintain coverage and effectively control risks in scenarios where standard conformal prediction fails.", "date": "2025-06-05", "categories": []} {"id": "2504.15360", "title": "Reliable Classification with Conformal Learning and Interval-Type 2 Fuzzy Sets", "abstract": "Classical machine learning classifiers tend to be overconfident can be unreliable outside of the laboratory benchmarks. Properly assessing the reliability of the output of the model per sample is instrumental for real-life scenarios where these systems are deployed. Because of this, different techniques have been employed to properly quantify the quality of prediction for a given model. These are most commonly Bayesian statistics and, more recently, conformal learning. Given a calibration set, conformal learning can produce outputs that are guaranteed to cover the target class with a desired significance level, and are more reliable than the standard confidence intervals used by Bayesian methods. In this work, we propose to use conformal learning with fuzzy rule-based systems in classification and show some metrics of their performance. Then, we discuss how the use of type 2 fuzzy sets can improve the quality of the output of the system compared to both fuzzy and crisp rules. Finally, we also discuss how the fine-tuning of the system can be adapted to improve the quality of the conformal prediction.", "date": "2025-04-21", "categories": []} {"id": "2511.04275", "title": "Online Conformal Inference with Retrospective Adjustment for Faster Adaptation to Distribution Shift", "abstract": "Conformal prediction has emerged as a powerful framework for constructing distribution-free prediction sets with guaranteed coverage assuming only the exchangeability assumption. However, this assumption is often violated in online environments where data distributions evolve over time. Several recent approaches have been proposed to address this limitation, but, typically, they slowly adapt to distribution shifts because they update predictions only in a forward manner, that is, they generate a prediction for a newly observed data point while previously computed predictions are not updated. In this paper, we propose a novel online conformal inference method with retrospective adjustment, which is designed to achieve faster adaptation to distributional shifts. Our method leverages regression approaches with efficient leave-one-out update formulas to retroactively adjust past predictions when new data arrive, thereby aligning the entire set of predictions with the most recent data distribution. Through extensive numerical studies performed on both synthetic and real-world data sets, we show that the proposed approach achieves faster coverage recalibration and improved statistical efficiency compared to existing online conformal prediction methods.", "date": "2025-11-06", "categories": []} {"id": "2505.24038", "title": "Conformal Object Detection by Sequential Risk Control", "abstract": "Recent advances in object detectors have led to their adoption for industrial uses. However, their deployment in safety-critical applications is hindered by the inherent lack of reliability of neural networks and the complex structure of object detection models. To address these challenges, we turn to Conformal Prediction, a post-hoc predictive uncertainty quantification procedure with statistical guarantees that are valid for any dataset size, without requiring prior knowledge on the model or data distribution. Our contribution is manifold. First, we formally define the problem of Conformal Object Detection (COD). We introduce a novel method, Sequential Conformal Risk Control (SeqCRC), that extends the statistical guarantees of Conformal Risk Control to two sequential tasks with two parameters, as required in the COD setting. Then, we present old and new loss functions and prediction sets suited to applying SeqCRC to different cases and certification requirements. Finally, we present a conformal toolkit for replication and further exploration of our method. Using this toolkit, we perform extensive experiments that validate our approach and emphasize trade-offs and other practical consequences.", "date": "2025-05-29", "categories": []} {"id": "2510.02809", "title": "Relevance-Aware Thresholding in Online Conformal Prediction for Time Series", "abstract": "Uncertainty quantification has received considerable interest in recent works in Machine Learning. In particular, Conformal Prediction (CP) gains ground in this field. For the case of time series, Online Conformal Prediction (OCP) becomes an option to address the problem of data distribution shift over time. Indeed, the idea of OCP is to update a threshold of some quantity (whether the miscoverage level or the quantile) based on the distribution observation. To evaluate the performance of OCP methods, two key aspects are typically considered: the coverage validity and the prediction interval width minimization. Recently, new OCP methods have emerged, offering long-run coverage guarantees and producing more informative intervals. However, during the threshold update step, most of these methods focus solely on the validity of the prediction intervals~--~that is, whether the ground truth falls inside or outside the interval~--~without accounting for their relevance. In this paper, we aim to leverage this overlooked aspect. Specifically, we propose enhancing the threshold update step by replacing the binary evaluation (inside/outside) with a broader class of functions that quantify the relevance of the prediction interval using the ground truth. This approach helps prevent abrupt threshold changes, potentially resulting in narrower prediction intervals. Indeed, experimental results on real-world datasets suggest that these functions can produce tighter intervals compared to existing OCP methods while maintaining coverage validity.", "date": "2025-10-03", "categories": []} {"id": "2510.26026", "title": "Conformal Prediction Beyond the Horizon: Distribution-Free Inference for Policy Evaluation", "abstract": "Reliable uncertainty quantification is crucial for reinforcement learning (RL) in high-stakes settings. We propose a unified conformal prediction framework for infinite-horizon policy evaluation that constructs distribution-free prediction intervals {for returns} in both on-policy and off-policy settings. Our method integrates distributional RL with conformal calibration, addressing challenges such as unobserved returns, temporal dependencies, and distributional shifts. We propose a modular pseudo-return construction based on truncated rollouts and a time-aware calibration strategy using experience replay and weighted subsampling. These innovations mitigate model bias and restore approximate exchangeability, enabling uncertainty quantification even under policy shifts. Our theoretical analysis provides coverage guarantees that account for model misspecification and importance weight estimation. Empirical results, including experiments in synthetic and benchmark environments like Mountain Car, show that our method significantly improves coverage and reliability over standard distributional RL baselines.", "date": "2025-10-29", "categories": []} {"id": "2307.13124", "title": "Conformal prediction for frequency-severity modeling", "abstract": "We present a model-agnostic framework for the construction of prediction intervals of insurance claims, with finite sample statistical guarantees, extending the technique of split conformal prediction to the domain of two-stage frequency-severity modeling. The framework effectiveness is showcased with simulated and real datasets using classical parametric models and contemporary machine learning methods. When the underlying severity model is a random forest, we extend the two-stage split conformal prediction algorithm, showing how the out-of-bag mechanism can be leveraged to eliminate the need for a calibration set in the conformal procedure.", "date": "2023-07-24", "categories": []} {"id": "2309.01778", "title": "CONFIDERAI: a novel CONFormal Interpretable-by-Design score function for Explainable and Reliable Artificial Intelligence", "abstract": "Everyday life is increasingly influenced by artificial intelligence, and there is no question that machine learning algorithms must be designed to be reliable and trustworthy for everyone. Specifically, computer scientists consider an artificial intelligence system safe and trustworthy if it fulfills five pillars: explainability, robustness, transparency, fairness, and privacy. In addition to these five, we propose a sixth fundamental aspect: conformity, that is, the probabilistic assurance that the system will behave as the machine learner expects. In this paper, we present a methodology to link conformal prediction with explainable machine learning by defining a new score function for rule-based classifiers that leverages rules predictive ability, the geometrical position of points within rules boundaries and the overlaps among rules as well, thanks to the definition of a geometrical rule similarity term. Furthermore, we address the problem of defining regions in the feature space where conformal guarantees are satisfied, by exploiting the definition of conformal critical set and showing how this set can be used to achieve new rules with improved performance on the target class. The overall methodology is tested with promising results on several datasets of real-world interest, such as domain name server tunneling detection or cardiovascular disease prediction.", "date": "2023-09-04", "categories": []} {"id": "2510.22931", "title": "Robust Uncertainty Quantification for Self-Evolving Large Language Models via Continual Domain Pretraining", "abstract": "Continual Learning (CL) is essential for enabling self-evolving large language models (LLMs) to adapt and remain effective amid rapid knowledge growth. Yet, despite its importance, little attention has been given to establishing statistical reliability guarantees for LLMs under CL, particularly in the setting of continual domain pretraining (CDP). Conformal Prediction (CP) has shown promise in offering correctness guarantees for LLMs, but it faces major challenges in CDP: testing data often stems from unknown or shifting domain distributions, under which CP may no longer provide valid guarantees. Moreover, when high coverage is required, CP can yield excessively large prediction sets for unanswerable queries, reducing informativeness. To address these challenges, we introduce an adaptive rejection and non-exchangeable CP framework. Our method first estimates the distribution of questions across domains in the test set using transformer-based clustering, then reweights or resamples the calibration data accordingly. Building on this, adaptive rejection CP allows the LLM to selectively abstain from answering when its confidence or competence shifts significantly. Extensive experiments demonstrate that our framework enhances both the effectiveness and reliability of CP under CDP scenarios. Our code is available at: https://anonymous.4open.science/r/CPCL-8C12/", "date": "2025-10-27", "categories": []} {"id": "2505.21658", "title": "STACI: Spatio-Temporal Aleatoric Conformal Inference", "abstract": "Fitting Gaussian Processes (GPs) provides interpretable aleatoric uncertainty quantification for estimation of spatio-temporal fields. Spatio-temporal deep learning models, while scalable, typically assume a simplistic independent covariance matrix for the response, failing to capture the underlying correlation structure. However, spatio-temporal GPs suffer from issues of scalability and various forms of approximation bias resulting from restrictive assumptions of the covariance kernel function. We propose STACI, a novel framework consisting of a variational Bayesian neural network approximation of non-stationary spatio-temporal GP along with a novel spatio-temporal conformal inference algorithm. STACI is highly scalable, taking advantage of GPU training capabilities for neural network models, and provides statistically valid prediction intervals for uncertainty quantification. STACI outperforms competing GPs and deep methods in accurately approximating spatio-temporal processes and we show it easily scales to datasets with millions of observations.", "date": "2025-05-27", "categories": []} {"id": "2507.10425", "title": "Non-exchangeable Conformal Prediction with Optimal Transport: Tackling Distribution Shifts with Unlabeled Data", "abstract": "Conformal prediction is a distribution-free uncertainty quantification method that has gained popularity in the machine learning community due to its finite-sample guarantees and ease of use. Its most common variant, dubbed split conformal prediction, is also computationally efficient as it boils down to collecting statistics of the model predictions on some calibration data not yet seen by the model. Nonetheless, these guarantees only hold if the calibration and test data are exchangeable, a condition that is difficult to verify and often violated in practice due to so-called distribution shifts. The literature is rife with methods to mitigate the loss in coverage in this non-exchangeable setting, but these methods require some prior information on the type of distribution shift to be expected at test time. In this work, we study this problem via a new perspective, through the lens of optimal transport, and show that it is possible to estimate the loss in coverage and mitigate arbitrary distribution shifts, offering a principled and broadly applicable solution.", "date": "2025-07-14", "categories": []} {"id": "2510.17543", "title": "Reliable Inference in Edge-Cloud Model Cascades via Conformal Alignment", "abstract": "Edge intelligence enables low-latency inference via compact on-device models, but assuring reliability remains challenging. We study edge-cloud cascades that must preserve conditional coverage: whenever the edge returns a prediction set, it should contain the true label with a user-specified probability, as if produced by the cloud model. We formalize conditional coverage with respect to the cloud predictive distribution, and introduce a conformal alignment-based (CAb) cascading mechanism that certifies this property with user control over the risk level. Our method casts escalation from edge to cloud models as a multiple-hypothesis testing (MHT) problem, tailoring conformal alignment (CA) to select which inputs can be safely handled at the edge. The proposed CAb model cascading method yields statistical guarantees on the average fraction of edge decisions that satisfy cloud-level conditional coverage. The procedure applies to arbitrary edge prediction sets, including variants of conformal prediction (CP), and exposes a tunable trade-off among coverage, deferral rate, and set size. Experiments on CIFAR-100 image classification and the TeleQnA question-answering (QA) benchmark show that the proposed CAb cascade maintains the target conditional coverage for edge predictions while substantially reducing offloading to the cloud and incurring modest increases in prediction-set size.", "date": "2025-10-20", "categories": []} {"id": "2507.02496", "title": "Online Conformal Prediction with Efficiency Guarantees", "abstract": "We study the problem of conformal prediction in a novel online framework that directly optimizes efficiency. In our problem, we are given a target miscoverage rate $\\alpha > 0$, and a time horizon $T$. On each day $t \\le T$ an algorithm must output an interval $I_t \\subseteq [0, 1]$, then a point $y_t \\in [0, 1]$ is revealed. The goal of the algorithm is to achieve coverage, that is, $y_t \\in I_t$ on (close to) a $(1 - \\alpha)$-fraction of days, while maintaining efficiency, that is, minimizing the average volume (length) of the intervals played. This problem is an online analogue to the problem of constructing efficient confidence intervals.\n We study this problem over arbitrary and exchangeable (random order) input sequences. For exchangeable sequences, we show that it is possible to construct intervals that achieve coverage $(1 - \\alpha) - o(1)$, while having length upper bounded by the best fixed interval that achieves coverage in hindsight. For arbitrary sequences however, we show that any algorithm that achieves a $\\mu$-approximation in average length compared to the best fixed interval achieving coverage in hindsight, must make a multiplicative factor more mistakes than $\\alpha T$, where the multiplicative factor depends on $\\mu$ and the aspect ratio of the problem. Our main algorithmic result is a matching algorithm that can recover all Pareto-optimal settings of $\\mu$ and number of mistakes. Furthermore, our algorithm is deterministic and therefore robust to an adaptive adversary.\n This gap between the exchangeable and arbitrary settings is in contrast to the classical online learning problem. In fact, we show that no single algorithm can simultaneously be Pareto-optimal for arbitrary sequences and optimal for exchangeable sequences. On the algorithmic side, we give an algorithm that achieves the near-optimal tradeoff between the two cases.", "date": "2025-07-03", "categories": []} {"id": "2509.25473", "title": "Conformal Prediction for Signal Temporal Logic Inference", "abstract": "Signal Temporal Logic (STL) inference seeks to extract human-interpretable rules from time-series data, but existing methods lack formal confidence guarantees for the inferred rules. Conformal prediction (CP) is a technique that can provide statistical correctness guarantees, but is typically applied as a post-training wrapper without improving model learning. Instead, we introduce an end-to-end differentiable CP framework for STL inference that enhances both reliability and interpretability of the resulting formulas. We introduce a robustness-based nonconformity score, embed a smooth CP layer directly into training, and employ a new loss function that simultaneously optimizes inference accuracy and CP prediction sets with a single term. Following training, an exact CP procedure delivers statistical guarantees for the learned STL formulas. Experiments on benchmark time-series tasks show that our approach reduces uncertainty in predictions (i.e., it achieves high coverage while reducing prediction set size), and improves accuracy (i.e., the number of misclassifications when using a fixed threshold) over state-of-the-art baselines.", "date": "2025-09-29", "categories": []} {"id": "2403.15527", "title": "Conformal online model aggregation", "abstract": "Conformal prediction equips machine learning models with a reasonable notion of uncertainty quantification without making strong distributional assumptions. It wraps around any prediction model and converts point predictions into set predictions with a predefined marginal coverage guarantee. However, conformal prediction only works if we fix the underlying machine learning model in advance. A relatively unaddressed issue in conformal prediction is that of model selection and/or aggregation: given a set of prediction models, which one should we conformalize? This paper suggests that instead of performing model selection, it can be prudent and practical to perform conformal set aggregation in an online, adaptive fashion. We propose a wrapper that takes in several conformal prediction sets (themselves wrapped around black-box prediction models), and outputs a single adaptively-combined prediction set. Our method, called conformal online model aggregation (COMA), is based on combining the prediction sets from several algorithms by weighted voting, and can be thought of as a sort of online stacking of the underlying conformal sets. As long as the input sets have (distribution-free) coverage guarantees, COMA retains coverage guarantees, under a negative correlation assumption between errors and weights. We verify that the assumption holds empirically in all settings considered. COMA is well-suited for decentralized or distributed settings, where different users may have different models, and are only willing to share their prediction sets for a new test point in a black-box fashion. As we demonstrate, it is also well-suited to settings with distribution drift and shift, where model selection can be imprudent.", "date": "2024-03-22", "categories": []} {"id": "2505.16115", "title": "A Generic Framework for Conformal Fairness", "abstract": "Conformal Prediction (CP) is a popular method for uncertainty quantification with machine learning models. While conformal prediction provides probabilistic guarantees regarding the coverage of the true label, these guarantees are agnostic to the presence of sensitive attributes within the dataset. In this work, we formalize \\textit{Conformal Fairness}, a notion of fairness using conformal predictors, and provide a theoretically well-founded algorithm and associated framework to control for the gaps in coverage between different sensitive groups. Our framework leverages the exchangeability assumption (implicit to CP) rather than the typical IID assumption, allowing us to apply the notion of Conformal Fairness to data types and tasks that are not IID, such as graph data. Experiments were conducted on graph and tabular datasets to demonstrate that the algorithm can control fairness-related gaps in addition to coverage aligned with theoretical expectations.", "date": "2025-05-22", "categories": []} {"id": "2510.13297", "title": "Federated Conditional Conformal Prediction via Generative Models", "abstract": "Conformal Prediction (CP) provides distribution-free uncertainty quantification by constructing prediction sets that guarantee coverage of the true labels. This reliability makes CP valuable for high-stakes federated learning scenarios such as multi-center healthcare. However, standard CP assumes i.i.d. data, which is violated in federated settings where client distributions differ substantially. Existing federated CP methods address this by maintaining marginal coverage on each client, but such guarantees often fail to reflect input-conditional uncertainty. In this work, we propose Federated Conditional Conformal Prediction (Fed-CCP) via generative models, which aims for conditional coverage that adapts to local data heterogeneity. Fed-CCP leverages generative models, such as normalizing flows or diffusion models, to approximate conditional data distributions without requiring the sharing of raw data. This enables each client to locally calibrate conformal scores that reflect its unique uncertainty, while preserving global consistency through federated aggregation. Experiments on real datasets demonstrate that Fed-CCP achieves more adaptive prediction sets.", "date": "2025-10-15", "categories": []} {"id": "2510.16166", "title": "Extending Prediction-Powered Inference through Conformal Prediction", "abstract": "Prediction-powered inference is a recent methodology for the safe use of black-box ML models to impute missing data, strengthening inference of statistical parameters. However, many applications require strong properties besides valid inference, such as privacy, robustness or validity under continuous distribution shifts; deriving prediction-powered methods with such guarantees is generally an arduous process, and has to be done case by case. In this paper, we resolve this issue by connecting prediction-powered inference with conformal prediction: by performing imputation through a calibrated conformal set-predictor, we attain validity while achieving additional guarantees in a natural manner. We instantiate our procedure for the inference of means, Z- and M-estimation, as well as e-values and e-value-based procedures. Furthermore, in the case of e-values, ours is the first general prediction-powered procedure that operates off-line. We demonstrate these advantages by applying our method on private and time-series data. Both tasks are nontrivial within the standard prediction-powered framework but become natural under our method.", "date": "2025-10-17", "categories": []} {"id": "2510.16376", "title": "Conformal Prediction in The Loop: A Feedback-Based Uncertainty Model for Trajectory Optimization", "abstract": "Conformal Prediction (CP) is a powerful statistical machine learning tool to construct uncertainty sets with coverage guarantees, which has fueled its extensive adoption in generating prediction regions for decision-making tasks, e.g., Trajectory Optimization (TO) in uncertain environments. However, existing methods predominantly employ a sequential scheme, where decisions rely unidirectionally on the prediction regions, and consequently the information from decision-making fails to be fed back to instruct CP. In this paper, we propose a novel Feedback-Based CP (Fb-CP) framework for shrinking-horizon TO with a joint risk constraint over the entire mission time. Specifically, a CP-based posterior risk calculation method is developed by fully leveraging the realized trajectories to adjust the posterior allowable risk, which is then allocated to future times to update prediction regions. In this way, the information in the realized trajectories is continuously fed back to the CP, enabling attractive feedback-based adjustments of the prediction regions and a provable online improvement in trajectory performance. Furthermore, we theoretically prove that such adjustments consistently maintain the coverage guarantees of the prediction regions, thereby ensuring provable safety. Additionally, we develop a decision-focused iterative risk allocation algorithm with theoretical convergence analysis for allocating the posterior allowable risk which closely aligns with Fb-CP. Furthermore, we extend the proposed method to handle distribution shift. The effectiveness and superiority of the proposed method are demonstrated through benchmark experiments.", "date": "2025-10-18", "categories": []} {"id": "2510.15233", "title": "Adaptive Individual Uncertainty under Out-Of-Distribution Shift with Expert-Routed Conformal Prediction", "abstract": "Reliable, informative, and individual uncertainty quantification (UQ) remains missing in current ML community. This hinders the effective application of AI/ML to risk-sensitive domains. Most methods either fail to provide coverage on new data, inflate intervals so broadly that they are not actionable, or assign uncertainties that do not track actual error, especially under a distribution shift. In high-stakes drug discovery, protein-ligand affinity (PLI) prediction is especially challenging as assay noise is heterogeneous, chemical space is imbalanced and large, and practical evaluations routinely involve distribution shift. In this work, we introduce a novel uncertainty quantification method, Trustworthy Expert Split-conformal with Scaled Estimation for Efficient Reliable Adaptive intervals (TESSERA), that provides per-sample uncertainty with reliable coverage guarantee, informative and adaptive prediction interval widths that track the absolute error. We evaluate on protein-ligand binding affinity prediction under both independent and identically distributed (i.i.d.) and scaffold-based out-of-distribution (OOD) splits, comparing against strong UQ baselines. TESSERA attains near-nominal coverage and the best coverage-width trade-off as measured by the Coverage-Width Criterion (CWC), while maintaining competitive adaptivity (lowest Area Under the Sparsification Error (AUSE)). Size-Stratified Coverage (SSC) further confirms that intervals are right-sized, indicating width increases when data are scarce or noisy, and remain tight when predictions are reliable. By unifying Mixture of Expert (MoE) diversity with conformal calibration, TESSERA delivers trustworthy, tight, and adaptive uncertainties that are well-suited to selective prediction and downstream decision-making in the drug-discovery pipeline and other applications.", "date": "2025-10-17", "categories": []} {"id": "2510.15824", "title": "Blackwell's Approachability for Sequential Conformal Inference", "abstract": "We study conformal inference in non-exchangeable environments through the lens of Blackwell's theory of approachability. We first recast adaptive conformal inference (ACI, Gibbs and Cand\\`es, 2021) as a repeated two-player vector-valued finite game and characterize attainable coverage--efficiency tradeoffs. We then construct coverage and efficiency objectives under potential restrictions on the adversary's play, and design a calibration-based approachability strategy to achieve these goals. The resulting algorithm enjoys strong theoretical guarantees and provides practical insights, though its computational burden may limit deployment in practice.", "date": "2025-10-17", "categories": []} {"id": "2501.18363", "title": "Exploring the Noise Robustness of Online Conformal Prediction", "abstract": "Conformal prediction is an emerging technique for uncertainty quantification that constructs prediction sets guaranteed to contain the true label with a predefined probability. Recent work develops online conformal prediction methods that adaptively construct prediction sets to accommodate distribution shifts. However, existing algorithms typically assume perfect label accuracy which rarely holds in practice. In this work, we investigate the robustness of online conformal prediction under uniform label noise with a known noise rate, in both constant and dynamic learning rate schedules. We show that label noise causes a persistent gap between the actual mis-coverage rate and the desired rate $\\alpha$, leading to either overestimated or underestimated coverage guarantees. To address this issue, we propose Noise Robust Online Conformal Prediction (dubbed NR-OCP) by updating the threshold with a novel robust pinball loss, which provides an unbiased estimate of clean pinball loss without requiring ground-truth labels. Our theoretical analysis shows that NR-OCP eliminates the coverage gap in both constant and dynamic learning rate schedules, achieving a convergence rate of $\\mathcal{O}(T^{-1/2})$ for both empirical and expected coverage errors under uniform label noise. Extensive experiments demonstrate the effectiveness of our method by achieving both precise coverage and improved efficiency.", "date": "2025-01-30", "categories": []} {"id": "2505.15437", "title": "Adaptive Set-Mass Calibration with Conformal Prediction", "abstract": "Reliable probabilities are critical in high-risk applications, yet common calibration criteria (confidence, class-wise) are only necessary for full distributional calibration, and post-hoc methods often lack distribution-free guarantees.\n We propose a set-based notion of calibration, cumulative mass calibration, and a corresponding empirical error measure: the Cumulative Mass Calibration Error (CMCE). We develop a new calibration procedure that starts with conformal prediction to obtain a set of labels that gives the desired coverage.\n We then instantiate two simple post-hoc calibrators: a mass normalization and a temperature scaling-based rule, tuned to the conformal constraint.\n On multi-class image benchmarks, especially with a large number of classes, our methods consistently improve CMCE and standard metrics (ECE, cw-ECE, MCE) over baselines, delivering a practical, scalable framework with theoretical guarantees.", "date": "2025-05-21", "categories": []} {"id": "2510.13037", "title": "Conformal Inference for Open-Set and Imbalanced Classification", "abstract": "This paper presents a conformal prediction method for classification in highly imbalanced and open-set settings, where there are many possible classes and not all may be represented in the data. Existing approaches require a finite, known label space and typically involve random sample splitting, which works well when there is a sufficient number of observations from each class. Consequently, they have two limitations: (i) they fail to provide adequate coverage when encountering new labels at test time, and (ii) they may become overly conservative when predicting previously seen labels. To obtain valid prediction sets in the presence of unseen labels, we compute and integrate into our predictions a new family of conformal p-values that can test whether a new data point belongs to a previously unseen class. We study these p-values theoretically, establishing their optimality, and uncover an intriguing connection with the classical Good--Turing estimator for the probability of observing a new species. To make more efficient use of imbalanced data, we also develop a selective sample splitting algorithm that partitions training and calibration data based on label frequency, leading to more informative predictions. Despite breaking exchangeability, this allows maintaining finite-sample guarantees through suitable re-weighting. With both simulated and real data, we demonstrate our method leads to prediction sets with valid coverage even in challenging open-set scenarios with infinite numbers of possible labels, and produces more informative predictions under extreme class imbalance.", "date": "2025-10-14", "categories": []} {"id": "2510.07185", "title": "Split Conformal Classification with Unsupervised Calibration", "abstract": "Methods for split conformal prediction leverage calibration samples to transform any prediction rule into a set-prediction rule that complies with a target coverage probability. Existing methods provide remarkably strong performance guarantees with minimal computational costs. However, they require to use calibration samples composed by labeled examples different to those used for training. This requirement can be highly inconvenient, as it prevents the use of all labeled examples for training and may require acquiring additional labels solely for calibration. This paper presents an effective methodology for split conformal prediction with unsupervised calibration for classification tasks. In the proposed approach, set-prediction rules are obtained using unsupervised calibration samples together with supervised training samples previously used to learn the classification rule. Theoretical and experimental results show that the presented methods can achieve performance comparable to that with supervised calibration, at the expenses of a moderate degradation in performance guarantees and computational efficiency.", "date": "2025-10-08", "categories": []} {"id": "2506.20898", "title": "Multi-model Online Conformal Prediction with Graph-Structured Feedback", "abstract": "Online conformal prediction has demonstrated its capability to construct a prediction set for each incoming data point that covers the true label with a predetermined probability. To cope with potential distribution shift, multi-model online conformal prediction has been introduced to select and leverage different models from a preselected candidate set. Along with the improved flexibility, the choice of the preselected set also brings challenges. A candidate set that includes a large number of models may increase the computational complexity. In addition, the inclusion of irrelevant models with poor performance may negatively impact the performance and lead to unnecessarily large prediction sets. To address these challenges, we propose a novel multi-model online conformal prediction algorithm that identifies a subset of effective models at each time step by collecting feedback from a bipartite graph, which is refined upon receiving new data. A model is then selected from this subset to construct the prediction set, resulting in reduced computational complexity and smaller prediction sets. Additionally, we demonstrate that using prediction set size as feedback, alongside model loss, can significantly improve efficiency by constructing smaller prediction sets while still satisfying the required coverage guarantee. The proposed algorithms are proven to ensure valid coverage and achieve sublinear regret. Experiments on real and synthetic datasets validate that the proposed methods construct smaller prediction sets and outperform existing multi-model online conformal prediction approaches.", "date": "2025-06-26", "categories": []} {"id": "2510.08075", "title": "Multi-Condition Conformal Selection", "abstract": "Selecting high-quality candidates from large-scale datasets is critically important in resource-constrained applications such as drug discovery, precision medicine, and the alignment of large language models. While conformal selection methods offer a rigorous solution with False Discovery Rate (FDR) control, their applicability is confined to single-threshold scenarios (i.e., y > c) and overlooks practical needs for multi-condition selection, such as conjunctive or disjunctive conditions. In this work, we propose the Multi-Condition Conformal Selection (MCCS) algorithm, which extends conformal selection to scenarios with multiple conditions. In particular, we introduce a novel nonconformity score with regional monotonicity for conjunctive conditions and a global Benjamini-Hochberg (BH) procedure for disjunctive conditions, thereby establishing finite-sample FDR control with theoretical guarantees. The integration of these components enables the proposed method to achieve rigorous FDR-controlled selection in various multi-condition environments. Extensive experiments validate the superiority of MCCS over baselines, its generalizability across diverse condition combinations, different real-world modalities, and multi-task scalability.", "date": "2025-10-09", "categories": []} {"id": "2510.11049", "title": "Conformal Inference for Time Series over Graphs", "abstract": "Trustworthy decision making in networked, dynamic environments calls for innovative uncertainty quantification substrates in predictive models for graph time series. Existing conformal prediction (CP) methods have been applied separately to multivariate time series and static graphs, but they either ignore the underlying graph topology or neglect temporal dynamics. To bridge this gap, here we develop a CP-based sequential prediction region framework tailored for graph time series. A key technical innovation is to leverage the graph structure and thus capture pairwise dependencies across nodes, while providing user-specified coverage guarantees on the predictive outcomes. We formally establish that our scheme yields an exponential shrinkage in the volume of the ellipsoidal prediction set relative to its graph-agnostic counterpart. Using real-world datasets, we demonstrate that the novel uncertainty quantification framework maintains desired empirical coverage while achieving markedly smaller (up to 80% reduction) prediction regions than existing approaches.", "date": "2025-10-13", "categories": []} {"id": "2510.08748", "title": "Conformal Risk Training: End-to-End Optimization of Conformal Risk Control", "abstract": "While deep learning models often achieve high predictive accuracy, their predictions typically do not come with any provable guarantees on risk or reliability, which are critical for deployment in high-stakes applications. The framework of conformal risk control (CRC) provides a distribution-free, finite-sample method for controlling the expected value of any bounded monotone loss function and can be conveniently applied post-hoc to any pre-trained deep learning model. However, many real-world applications are sensitive to tail risks, as opposed to just expected loss. In this work, we develop a method for controlling the general class of Optimized Certainty-Equivalent (OCE) risks, a broad class of risk measures which includes as special cases the expected loss (generalizing the original CRC method) and common tail risks like the conditional value-at-risk (CVaR). Furthermore, standard post-hoc CRC can degrade average-case performance due to its lack of feedback to the model. To address this, we introduce \"conformal risk training,\" an end-to-end approach that differentiates through conformal OCE risk control during model training or fine-tuning. Our method achieves provable risk guarantees while demonstrating significantly improved average-case performance over post-hoc approaches on applications to controlling classifiers' false negative rate and controlling financial risk in battery storage operation.", "date": "2025-10-09", "categories": []} {"id": "2307.12472", "title": "Model-free generalized fiducial inference", "abstract": "Conformal prediction (CP) was developed to provide finite-sample probabilistic prediction guarantees. While CP algorithms are a relatively general-purpose approach to uncertainty quantification, with finite-sample guarantees, they lack versatility. Namely, the CP approach does not {\\em prescribe} how to quantify the degree to which a data set provides evidence in support of (or against) an arbitrary event from a general class of events. In this paper, tools are offered from imprecise probability theory to build a formal connection between CP and generalized fiducial (GF) inference. These new insights establish a more general inferential lens from which CP can be understood, and demonstrate the pragmatism of fiducial ideas. The formal connection establishes a context in which epistemically-derived GF probability matches aleatoric/frequentist probability. Beyond this fact, it is illustrated how tools from imprecise probability theory, namely lower and upper probability functions, can be applied in the context of the imprecise GF distribution to provide posterior-like, prescriptive inference that is not possible within the CP framework alone. In addition to the primary CP generalization that is contributed, fundamental connections are synthesized between this new model-free GF and three other areas of contemporary research: nonparametric predictive inference (NPI), conformal predictive systems/distributions, and inferential models (IMs).", "date": "2023-07-24", "categories": []} {"id": "2510.06181", "title": "Conformalized Gaussian processes for online uncertainty quantification over graphs", "abstract": "Uncertainty quantification (UQ) over graphs arises in a number of safety-critical applications in network science. The Gaussian process (GP), as a classical Bayesian framework for UQ, has been developed to handle graph-structured data by devising topology-aware kernel functions. However, such GP-based approaches are limited not only by the prohibitive computational complexity, but also the strict modeling assumptions that might yield poor coverage, especially with labels arriving on the fly. To effect scalability, we devise a novel graph-aware parametric GP model by leveraging the random feature (RF)-based kernel approximation, which is amenable to efficient recursive Bayesian model updates. To further allow for adaptivity, an ensemble of graph-aware RF-based scalable GPs have been leveraged, with per-GP weight adapted to data arriving incrementally. To ensure valid coverage with robustness to model mis-specification, we wed the GP-based set predictors with the online conformal prediction framework, which post-processes the prediction sets using adaptive thresholds. Experimental results the proposed method yields improved coverage and efficient prediction sets over existing baselines by adaptively ensembling the GP models and setting the key threshold parameters in CP.", "date": "2025-10-07", "categories": []} {"id": "2502.16736", "title": "Adaptive Conformal Guidance for Learning under Uncertainty", "abstract": "Learning with guidance has proven effective across a wide range of machine learning systems. Guidance may, for example, come from annotated datasets in supervised learning, pseudo-labels in semi-supervised learning, and expert demonstration policies in reinforcement learning. However, guidance signals can be noisy due to domain shifts and limited data availability and may not generalize well. Blindly trusting such signals when they are noisy, incomplete, or misaligned with the target domain can lead to degraded performance. To address these challenges, we propose Adaptive Conformal Guidance (AdaConG), a simple yet effective approach that dynamically modulates the influence of guidance signals based on their associated uncertainty, quantified via split conformal prediction (CP). By adaptively adjusting to guidance uncertainty, AdaConG enables models to reduce reliance on potentially misleading signals and enhance learning performance. We validate AdaConG across diverse tasks, including knowledge distillation, semi-supervised image classification, gridworld navigation, and autonomous driving. Experimental results demonstrate that AdaConG improves performance and robustness under imperfect guidance, e.g., in gridworld navigation, it accelerates convergence and achieves over $6\\times$ higher rewards than the best-performing baseline. These results highlight AdaConG as a broadly applicable solution for learning under uncertainty.", "date": "2025-02-23", "categories": []} {"id": "2509.25444", "title": "Neural Optimal Transport Meets Multivariate Conformal Prediction", "abstract": "We propose a framework for conditional vector quantile regression (CVQR) that combines neural optimal transport with amortized optimization, and apply it to multivariate conformal prediction. Classical quantile regression does not extend naturally to multivariate responses, while existing approaches often ignore the geometry of joint distributions. Our method parametrizes the conditional vector quantile function as the gradient of a convex potential implemented by an input-convex neural network, ensuring monotonicity and uniform ranks. To reduce the cost of solving high-dimensional variational problems, we introduced amortized optimization of the dual potentials, yielding efficient training and faster inference. We then exploit the induced multivariate ranks for conformal prediction, constructing distribution-free predictive regions with finite-sample validity. Unlike coordinatewise methods, our approach adapts to the geometry of the conditional distribution, producing tighter and more informative regions. Experiments on benchmark datasets show improved coverage-efficiency trade-offs compared to baselines, highlighting the benefits of integrating neural optimal transport with conformal prediction.", "date": "2025-09-29", "categories": []} {"id": "2509.25692", "title": "Annotation-Efficient Active Test-Time Adaptation with Conformal Prediction", "abstract": "Active Test-Time Adaptation (ATTA) improves model robustness under domain shift by selectively querying human annotations at deployment, but existing methods use heuristic uncertainty measures and suffer from low data selection efficiency, wasting human annotation budget. We propose Conformal Prediction Active TTA (CPATTA), which first brings principled, coverage-guaranteed uncertainty into ATTA. CPATTA employs smoothed conformal scores with a top-K certainty measure, an online weight-update algorithm driven by pseudo coverage, a domain-shift detector that adapts human supervision, and a staged update scheme balances human-labeled and model-labeled data. Extensive experiments demonstrate that CPATTA consistently outperforms the state-of-the-art ATTA methods by around 5% in accuracy. Our code and datasets are available at https://github.com/tingyushi/CPATTA.", "date": "2025-09-30", "categories": []} {"id": "2503.03659", "title": "Conformal prediction of future insurance claims in the regression problem", "abstract": "In the current insurance literature, prediction of insurance claims in the regression problem is often performed with a statistical model. This model-based approach may potentially suffer from several drawbacks: (i) model misspecification, (ii) selection effect, and (iii) lack of finite-sample validity. This article addresses these three issues simultaneously by employing conformal prediction -- a general machine learning strategy for valid predictions. The proposed method is both model-free and tuning-parameter-free. It also guarantees finite-sample validity at a pre-assigned coverage probability level. Examples, based on both simulated and real data, are provided to demonstrate the excellent performance of the proposed method and its applications in insurance, especially regarding meeting the solvency capital requirement of European insurance regulation, Solvency II.", "date": "2025-03-05", "categories": []} {"id": "2507.17796", "title": "CoCAI: Copula-based Conformal Anomaly Identification for Multivariate Time-Series", "abstract": "We propose a novel framework that harnesses the power of generative artificial intelligence and copula-based modeling to address two critical challenges in multivariate time-series analysis: delivering accurate predictions and enabling robust anomaly detection. Our method, Copula-based Conformal Anomaly Identification for Multivariate Time-Series (CoCAI), leverages a diffusion-based model to capture complex dependencies within the data, enabling high quality forecasting. The model's outputs are further calibrated using a conformal prediction technique, yielding predictive regions which are statistically valid, i.e., cover the true target values with a desired confidence level. Starting from these calibrated forecasts, robust outlier detection is performed by combining dimensionality reduction techniques with copula-based modeling, providing a statistically grounded anomaly score. CoCAI benefits from an offline calibration phase that allows for minimal overhead during deployment and delivers actionable results rooted in established theoretical foundations. Empirical tests conducted on real operational data derived from water distribution and sewerage systems confirm CoCAI's effectiveness in accurately forecasting target sequences of data and in identifying anomalous segments within them.", "date": "2025-07-23", "categories": []} {"id": "2404.15274", "title": "Metric-Guided Conformal Bounds for Probabilistic Image Reconstruction", "abstract": "Modern deep learning reconstruction algorithms generate impressively realistic scans from sparse inputs, but can often produce significant inaccuracies. This makes it difficult to provide statistically guaranteed claims about the true state of a subject from scans reconstructed by these algorithms. In this study, we propose a framework for computing provably valid prediction bounds on claims derived from probabilistic black-box image reconstruction algorithms. The key insights behind our framework are to represent reconstructed scans with a derived clinical metric of interest, and to calibrate bounds on the ground truth metric with conformal prediction (CP) using a prior calibration dataset. These bounds convey interpretable feedback about the subject's state, and can also be used to retrieve nearest-neighbor reconstructed scans for visual inspection. We demonstrate the utility of this framework on sparse-view computed tomography (CT) for fat mass quantification and radiotherapy planning tasks. Results show that our framework produces bounds with better semantical interpretation than conventional pixel-based bounding approaches. Furthermore, we can flag dangerous outlier reconstructions that look plausible but have statistically unlikely metric values.", "date": "2024-04-23", "categories": []} {"id": "2509.21660", "title": "A Systematic Review of Conformal Inference Procedures for Treatment Effect Estimation: Methods and Challenges", "abstract": "Treatment effect estimation is essential for informed decision-making in many fields such as healthcare, economics, and public policy. While flexible machine learning models have been widely applied for estimating heterogeneous treatment effects, quantifying the inherent uncertainty of their point predictions remains an issue. Recent advancements in conformal prediction address this limitation by allowing for inexpensive computation, as well as distribution shifts, while still providing frequentist, finite-sample coverage guarantees under minimal assumptions for any point-predictor model. This advancement holds significant potential for improving decision-making in especially high-stakes environments. In this work, we perform a systematic review regarding conformal prediction methods for treatment effect estimation and provide for both the necessary theoretical background. Through a systematic filtering process, we select and analyze eleven key papers, identifying and describing current state-of-the-art methods in this area. Based on our findings, we propose directions for future research.", "date": "2025-09-25", "categories": []} {"id": "2509.21955", "title": "Learnable Conformal Prediction with Context-Aware Nonconformity Functions for Robotic Planning and Perception", "abstract": "Deep learning models in robotics often output point estimates with poorly calibrated confidences, offering no native mechanism to quantify predictive reliability under novel, noisy, or out-of-distribution inputs. Conformal prediction (CP) addresses this gap by providing distribution-free coverage guarantees, yet its reliance on fixed nonconformity scores ignores context and can yield intervals that are overly conservative or unsafe. We address this with Learnable Conformal Prediction (LCP), which replaces fixed scores with a lightweight neural function that leverages geometric, semantic, and task-specific features to produce context-aware uncertainty sets.\n LCP maintains CP's theoretical guarantees while reducing prediction set sizes by 18% in classification, tightening detection intervals by 52%, and improving path planning safety from 72% to 91% success with minimal overhead. Across three robotic tasks on seven benchmarks, LCP consistently outperforms Standard CP and ensemble baselines. In classification on CIFAR-100 and ImageNet, it achieves smaller set sizes (4.7-9.9% reduction) at target coverage. For object detection on COCO, BDD100K, and Cityscapes, it produces 46-54% tighter bounding boxes. In path planning through cluttered environments, it improves success to 91.5% with only 4.5% path inflation, compared to 12.2% for Standard CP.\n The method is lightweight (approximately 4.8% runtime overhead, 42 KB memory) and supports online adaptation, making it well suited to resource-constrained autonomous systems. Hardware evaluation shows LCP adds less than 1% memory and 15.9% inference overhead, yet sustains 39 FPS on detection tasks while being 7.4 times more energy-efficient than ensembles.", "date": "2025-09-26", "categories": []} {"id": "2509.22529", "title": "Smoothing-Based Conformal Prediction for Balancing Efficiency and Interpretability", "abstract": "Conformal Prediction (CP) is a distribution-free framework for constructing statistically rigorous prediction sets. While popular variants such as CD-split improve CP's efficiency, they often yield prediction sets composed of multiple disconnected subintervals, which are difficult to interpret. In this paper, we propose SCD-split, which incorporates smoothing operations into the CP framework. Such smoothing operations potentially help merge the subintervals, thus leading to interpretable prediction sets. Experimental results on both synthetic and real-world datasets demonstrate that SCD-split balances the interval length and the number of disconnected subintervals. Theoretically, under specific conditions, SCD-split provably reduces the number of disconnected subintervals while maintaining comparable coverage guarantees and interval length compared with CD-split.", "date": "2025-09-26", "categories": []} {"id": "2502.07255", "title": "Beyond Confidence: Adaptive Abstention in Dual-Threshold Conformal Prediction for Autonomous System Perception", "abstract": "Safety-critical perception systems require both reliable uncertainty quantification and principled abstention mechanisms to maintain safety under diverse operational conditions. We present a novel dual-threshold conformalization framework that provides statistically-guaranteed uncertainty estimates while enabling selective prediction in high-risk scenarios. Our approach uniquely combines a conformal threshold ensuring valid prediction sets with an abstention threshold optimized through ROC analysis, providing distribution-free coverage guarantees (>= 1 - alpha) while identifying unreliable predictions. Through comprehensive evaluation on CIFAR-100, ImageNet1K, and ModelNet40 datasets, we demonstrate superior robustness across camera and LiDAR modalities under varying environmental perturbations. The framework achieves exceptional detection performance (AUC: 0.993 to 0.995) under severe conditions while maintaining high coverage (>90.0%) and enabling adaptive abstention (13.5% to 63.4% +/- 0.5) as environmental severity increases. For LiDAR-based perception, our approach demonstrates particularly strong performance, maintaining robust coverage (>84.5%) while appropriately abstaining from unreliable predictions. Notably, the framework shows remarkable stability under heavy perturbations, with detection performance (AUC: 0.995 +/- 0.001) significantly outperforming existing methods across all modalities. Our unified approach bridges the gap between theoretical guarantees and practical deployment needs, offering a robust solution for safety-critical autonomous systems operating in challenging real-world conditions.", "date": "2025-02-11", "categories": []} {"id": "2503.21251", "title": "Dual-Splitting Conformal Prediction for Multi-Step Time Series Forecasting", "abstract": "Time series forecasting is crucial for applications like resource scheduling and risk management, where multi-step predictions provide a comprehensive view of future trends. Uncertainty Quantification (UQ) is a mainstream approach for addressing forecasting uncertainties, with Conformal Prediction (CP) gaining attention due to its model-agnostic nature and statistical guarantees. However, most variants of CP are designed for single-step predictions and face challenges in multi-step scenarios, such as reliance on real-time data and limited scalability. This highlights the need for CP methods specifically tailored to multi-step forecasting. We propose the Dual-Splitting Conformal Prediction (DSCP) method, a novel CP approach designed to capture inherent dependencies within time-series data for multi-step forecasting. Experimental results on real-world datasets from four different domains demonstrate that the proposed DSCP significantly outperforms existing CP variants in terms of the Winkler Score, achieving a performance improvement of up to 23.59% compared to state-of-the-art methods. Furthermore, we deployed the DSCP approach for renewable energy generation and IT load forecasting in power management of a real-world trajectory-based application, achieving an 11.25% reduction in carbon emissions through predictive optimization of data center operations and controls.", "date": "2025-03-27", "categories": []} {"id": "2509.15349", "title": "Probabilistic Conformal Coverage Guarantees in Small-Data Settings", "abstract": "Conformal prediction provides distribution-free prediction sets with guaranteed marginal coverage. However, in split conformal prediction this guarantee is training-conditional only in expectation: across many calibration draws, the average coverage equals the nominal level, but the realized coverage for a single calibration set may vary substantially. This variance undermines effective risk control in practical applications. Here we introduce the Small Sample Beta Correction (SSBC), a plug-and-play adjustment to the conformal significance level that leverages the exact finite-sample distribution of conformal coverage to provide probabilistic guarantees, ensuring that with user-defined probability over the calibration draw, the deployed predictor achieves at least the desired coverage.", "date": "2025-09-18", "categories": []} {"id": "2509.15485", "title": "mucAI at BAREC Shared Task 2025: Towards Uncertainty Aware Arabic Readability Assessment", "abstract": "We present a simple, model-agnostic post-processing technique for fine-grained Arabic readability classification in the BAREC 2025 Shared Task (19 ordinal levels). Our method applies conformal prediction to generate prediction sets with coverage guarantees, then computes weighted averages using softmax-renormalized probabilities over the conformal sets. This uncertainty-aware decoding improves Quadratic Weighted Kappa (QWK) by reducing high-penalty misclassifications to nearer levels. Our approach shows consistent QWK improvements of 1-3 points across different base models. In the strict track, our submission achieves QWK scores of 84.9\\%(test) and 85.7\\% (blind test) for sentence level, and 73.3\\% for document level. For Arabic educational assessment, this enables human reviewers to focus on a handful of plausible levels, combining statistical guarantees with practical usability.", "date": "2025-09-18", "categories": []} {"id": "2509.15120", "title": "Efficient Conformal Prediction for Regression Models under Label Noise", "abstract": "In high-stakes scenarios, such as medical imaging applications, it is critical to equip the predictions of a regression model with reliable confidence intervals. Recently, Conformal Prediction (CP) has emerged as a powerful statistical framework that, based on a labeled calibration set, generates intervals that include the true labels with a pre-specified probability. In this paper, we address the problem of applying CP for regression models when the calibration set contains noisy labels. We begin by establishing a mathematically grounded procedure for estimating the noise-free CP threshold. Then, we turn it into a practical algorithm that overcomes the challenges arising from the continuous nature of the regression problem. We evaluate the proposed method on two medical imaging regression datasets with Gaussian label noise. Our method significantly outperforms the existing alternative, achieving performance close to the clean-label setting.", "date": "2025-09-18", "categories": []} {"id": "2509.13717", "title": "A Conformal Prediction Framework for Uncertainty Quantification in Physics-Informed Neural Networks", "abstract": "Physics-Informed Neural Networks (PINNs) have emerged as a powerful framework for solving PDEs, yet existing uncertainty quantification (UQ) approaches for PINNs generally lack rigorous statistical guarantees. In this work, we bridge this gap by introducing a distribution-free conformal prediction (CP) framework for UQ in PINNs. This framework calibrates prediction intervals by constructing nonconformity scores on a calibration set, thereby yielding distribution-free uncertainty estimates with rigorous finite-sample coverage guarantees for PINNs. To handle spatial heteroskedasticity, we further introduce local conformal quantile estimation, enabling spatially adaptive uncertainty bands while preserving theoretical guarantee. Through systematic evaluations on typical PDEs (damped harmonic oscillator, Poisson, Allen-Cahn, and Helmholtz equations) and comprehensive testing across multiple uncertainty metrics, our results demonstrate that the proposed framework achieves reliable calibration and locally adaptive uncertainty intervals, consistently outperforming heuristic UQ approaches. By bridging PINNs with distribution-free UQ, this work introduces a general framework that not only enhances calibration and reliability, but also opens new avenues for uncertainty-aware modeling of complex PDE systems.", "date": "2025-09-17", "categories": []} {"id": "2502.00818", "title": "Error-quantified Conformal Inference for Time Series", "abstract": "Uncertainty quantification in time series prediction is challenging due to the temporal dependence and distribution shift on sequential data. Conformal inference provides a pivotal and flexible instrument for assessing the uncertainty of machine learning models through prediction sets. Recently, a series of online conformal inference methods updated thresholds of prediction sets by performing online gradient descent on a sequence of quantile loss functions. A drawback of such methods is that they only use the information of revealed non-conformity scores via miscoverage indicators but ignore error quantification, namely the distance between the non-conformity score and the current threshold. To accurately leverage the dynamic of miscoverage error, we propose \\textit{Error-quantified Conformal Inference} (ECI) by smoothing the quantile loss function. ECI introduces a continuous and adaptive feedback scale with the miscoverage error, rather than simple binary feedback in existing methods. We establish a long-term coverage guarantee for ECI under arbitrary dependence and distribution shift. The extensive experimental results show that ECI can achieve valid miscoverage control and output tighter prediction sets than other baselines.", "date": "2025-02-02", "categories": []} {"id": "2509.04623", "title": "Split Conformal Prediction in the Function Space with Neural Operators", "abstract": "Uncertainty quantification for neural operators remains an open problem in the infinite-dimensional setting due to the lack of finite-sample coverage guarantees over functional outputs. While conformal prediction offers finite-sample guarantees in finite-dimensional spaces, it does not directly extend to function-valued outputs. Existing approaches (Gaussian processes, Bayesian neural networks, and quantile-based operators) require strong distributional assumptions or yield conservative coverage. This work extends split conformal prediction to function spaces following a two step method. We first establish finite-sample coverage guarantees in a finite-dimensional space using a discretization map in the output function space. Then these guarantees are lifted to the function-space by considering the asymptotic convergence as the discretization is refined. To characterize the effect of resolution, we decompose the conformal radius into discretization, calibration, and misspecification components. This decomposition motivates a regression-based correction to transfer calibration across resolutions. Additionally, we propose two diagnostic metrics (conformal ensemble score and internal agreement) to quantify forecast degradation in autoregressive settings. Empirical results show that our method maintains calibrated coverage with less variation under resolution shifts and achieves better coverage in super-resolution tasks.", "date": "2025-09-04", "categories": []} {"id": "2509.04733", "title": "CoVeR: Conformal Calibration for Versatile and Reliable Autoregressive Next-Token Prediction", "abstract": "Autoregressive pre-trained models combined with decoding methods have achieved impressive performance on complex reasoning tasks. While mainstream decoding strategies such as beam search can generate plausible candidate sets, they often lack provable coverage guarantees, and struggle to effectively balance search efficiency with the need for versatile trajectories, particularly those involving long-tail sequences that are essential in certain real-world applications. To address these limitations, we propose \\textsc{CoVeR}, a novel model-free decoding strategy wihtin the conformal prediction framework that simultaneously maintains a compact search space and ensures high coverage probability over desirable trajectories. Theoretically, we establish a PAC-style generalization bound, guaranteeing that \\textsc{CoVeR} asymptotically achieves a coverage rate of at least $1 - \\alpha$ for any target level $\\alpha \\in (0,1)$.", "date": "2025-09-05", "categories": []} {"id": "2509.01338", "title": "Conformal Predictive Monitoring for Multi-Modal Scenarios", "abstract": "We consider the problem of quantitative predictive monitoring (QPM) of stochastic systems, i.e., predicting at runtime the degree of satisfaction of a desired temporal logic property from the current state of the system. Since computational efficiency is key to enable timely intervention against predicted violations, several state-of-the-art QPM approaches rely on fast machine-learning surrogates to provide prediction intervals for the satisfaction values, using conformal inference to offer statistical guarantees. However, these QPM methods suffer when the monitored agent exhibits multi-modal dynamics, whereby certain modes may yield high satisfaction values while others critically violate the property. Existing QPM methods are mode-agnostic and so would yield overly conservative and uninformative intervals that lack meaningful mode-specific satisfaction information. To address this problem, we present GenQPM, a method that leverages deep generative models, specifically score-based diffusion models, to reliably approximate the probabilistic and multi-modal system dynamics without requiring explicit model access. GenQPM employs a mode classifier to partition the predicted trajectories by dynamical mode. For each mode, we then apply conformal inference to produce statistically valid, mode-specific prediction intervals. We demonstrate the effectiveness of GenQPM on a benchmark of agent navigation and autonomous driving tasks, resulting in prediction intervals that are significantly more informative (less conservative) than mode-agnostic baselines.", "date": "2025-09-01", "categories": []} {"id": "2505.04608", "title": "WATCH: Adaptive Monitoring for AI Deployments via Weighted-Conformal Martingales", "abstract": "Responsibly deploying artificial intelligence (AI) / machine learning (ML) systems in high-stakes settings arguably requires not only proof of system reliability, but also continual, post-deployment monitoring to quickly detect and address any unsafe behavior. Methods for nonparametric sequential testing -- especially conformal test martingales (CTMs) and anytime-valid inference -- offer promising tools for this monitoring task. However, existing approaches are restricted to monitoring limited hypothesis classes or ``alarm criteria'' (e.g., detecting data shifts that violate certain exchangeability or IID assumptions), do not allow for online adaptation in response to shifts, and/or cannot diagnose the cause of degradation or alarm. In this paper, we address these limitations by proposing a weighted generalization of conformal test martingales (WCTMs), which lay a theoretical foundation for online monitoring for any unexpected changepoints in the data distribution while controlling false-alarms. For practical applications, we propose specific WCTM algorithms that adapt online to mild covariate shifts (in the marginal input distribution), quickly detect harmful shifts, and diagnose those harmful shifts as concept shifts (in the conditional label distribution) or extreme (out-of-support) covariate shifts that cannot be easily adapted to. On real-world datasets, we demonstrate improved performance relative to state-of-the-art baselines.", "date": "2025-05-07", "categories": []} {"id": "2508.14266", "title": "Effect of Data Augmentation on Conformal Prediction for Diabetic Retinopathy", "abstract": "The clinical deployment of deep learning models for high-stakes tasks such as diabetic retinopathy (DR) grading requires demonstrable reliability. While models achieve high accuracy, their clinical utility is limited by a lack of robust uncertainty quantification. Conformal prediction (CP) offers a distribution-free framework to generate prediction sets with statistical guarantees of coverage. However, the interaction between standard training practices like data augmentation and the validity of these guarantees is not well understood. In this study, we systematically investigate how different data augmentation strategies affect the performance of conformal predictors for DR grading. Using the DDR dataset, we evaluate two backbone architectures -- ResNet-50 and a Co-Scale Conv-Attentional Transformer (CoaT) -- trained under five augmentation regimes: no augmentation, standard geometric transforms, CLAHE, Mixup, and CutMix. We analyze the downstream effects on conformal metrics, including empirical coverage, average prediction set size, and correct efficiency. Our results demonstrate that sample-mixing strategies like Mixup and CutMix not only improve predictive accuracy but also yield more reliable and efficient uncertainty estimates. Conversely, methods like CLAHE can negatively impact model certainty. These findings highlight the need to co-design augmentation strategies with downstream uncertainty quantification in mind to build genuinely trustworthy AI systems for medical imaging.", "date": "2025-08-19", "categories": []} {"id": "2502.16336", "title": "Rectifying Conformity Scores for Better Conditional Coverage", "abstract": "We present a new method for generating confidence sets within the split conformal prediction framework. Our method performs a trainable transformation of any given conformity score to improve conditional coverage while ensuring exact marginal coverage. The transformation is based on an estimate of the conditional quantile of conformity scores. The resulting method is particularly beneficial for constructing adaptive confidence sets in multi-output problems where standard conformal quantile regression approaches have limited applicability. We develop a theoretical bound that captures the influence of the accuracy of the quantile estimate on the approximate conditional validity, unlike classical bounds for conformal prediction methods that only offer marginal coverage. We experimentally show that our method is highly adaptive to the local data structure and outperforms existing methods in terms of conditional coverage, improving the reliability of statistical inference in various applications.", "date": "2025-02-22", "categories": []} {"id": "2508.13288", "title": "Hierarchical Conformal Classification", "abstract": "Conformal prediction (CP) is a powerful framework for quantifying uncertainty in machine learning models, offering reliable predictions with finite-sample coverage guarantees. When applied to classification, CP produces a prediction set of possible labels that is guaranteed to contain the true label with high probability, regardless of the underlying classifier. However, standard CP treats classes as flat and unstructured, ignoring domain knowledge such as semantic relationships or hierarchical structure among class labels. This paper presents hierarchical conformal classification (HCC), an extension of CP that incorporates class hierarchies into both the structure and semantics of prediction sets. We formulate HCC as a constrained optimization problem whose solutions yield prediction sets composed of nodes at different levels of the hierarchy, while maintaining coverage guarantees. To address the combinatorial nature of the problem, we formally show that a much smaller, well-structured subset of candidate solutions suffices to ensure coverage while upholding optimality. An empirical evaluation on three new benchmarks consisting of audio, image, and text data highlights the advantages of our approach, and a user study shows that annotators significantly prefer hierarchical over flat prediction sets.", "date": "2025-08-18", "categories": []} {"id": "2508.13838", "title": "Online Conformal Selection with Accept-to-Reject Changes", "abstract": "Selecting a subset of promising candidates from a large pool is crucial across various scientific and real-world applications. Conformal selection offers a distribution-free and model-agnostic framework for candidate selection with uncertainty quantification. While effective in offline settings, its application to online scenarios, where data arrives sequentially, poses challenges. Notably, conformal selection permits the deselection of previously selected candidates, which is incompatible with applications requiring irreversible selection decisions. This limitation is particularly evident in resource-intensive sequential processes, such as drug discovery, where advancing a compound to subsequent stages renders reversal impractical. To address this issue, we extend conformal selection to an online Accept-to-Reject Changes (ARC) procedure: non-selected data points can be reconsidered for selection later, and once a candidate is selected, the decision is irreversible. Specifically, we propose a novel conformal selection method, Online Conformal Selection with Accept-to-Reject Changes (dubbed OCS-ARC), which incorporates online Benjamini-Hochberg procedure into the candidate selection process. We provide theoretical guarantees that OCS-ARC controls the false discovery rate (FDR) at or below the nominal level at any timestep under both i.i.d. and exchangeable data assumptions. Additionally, we theoretically show that our approach naturally extends to multivariate response settings. Extensive experiments on synthetic and real-world datasets demonstrate that OCS-ARC significantly improves selection power over the baseline while maintaining valid FDR control across all examined timesteps.", "date": "2025-08-19", "categories": []} {"id": "2508.12905", "title": "TCUQ: Single-Pass Uncertainty Quantification from Temporal Consistency with Streaming Conformal Calibration for TinyML", "abstract": "We introduce TCUQ, a single pass, label free uncertainty monitor for streaming TinyML that converts short horizon temporal consistency captured via lightweight signals on posteriors and features into a calibrated risk score with an O(W ) ring buffer and O(1) per step updates. A streaming conformal layer turns this score into a budgeted accept/abstain rule, yielding calibrated behavior without online labels or extra forward passes. On microcontrollers, TCUQ fits comfortably on kilobyte scale devices and reduces footprint and latency versus early exit and deep ensembles (typically about 50 to 60% smaller and about 30 to 45% faster), while methods of similar accuracy often run out of memory. Under corrupted in distribution streams, TCUQ improves accuracy drop detection by 3 to 7 AUPRC points and reaches up to 0.86 AUPRC at high severities; for failure detection it attains up to 0.92 AUROC. These results show that temporal consistency, coupled with streaming conformal calibration, provides a practical and resource efficient foundation for on device monitoring in TinyML.", "date": "2025-08-18", "categories": []} {"id": "2508.11345", "title": "Conformal Prediction Meets Long-tail Classification", "abstract": "Conformal Prediction (CP) is a popular method for uncertainty quantification that converts a pretrained model's point prediction into a prediction set, with the set size reflecting the model's confidence. Although existing CP methods are guaranteed to achieve marginal coverage, they often exhibit imbalanced coverage across classes under long-tail label distributions, tending to over cover the head classes at the expense of under covering the remaining tail classes. This under coverage is particularly concerning, as it undermines the reliability of the prediction sets for minority classes, even with coverage ensured on average. In this paper, we propose the Tail-Aware Conformal Prediction (TACP) method to mitigate the under coverage of the tail classes by utilizing the long-tail structure and narrowing the head-tail coverage gap. Theoretical analysis shows that it consistently achieves a smaller head-tail coverage gap than standard methods. To further improve coverage balance across all classes, we introduce an extension of TACP: soft TACP (sTACP) via a reweighting mechanism. The proposed framework can be combined with various non-conformity scores, and experiments on multiple long-tail benchmark datasets demonstrate the effectiveness of our methods.", "date": "2025-08-15", "categories": []} {"id": "2508.10022", "title": "Conformal P-Value in Multiple-Choice Question Answering Tasks with Provable Risk Control", "abstract": "This study introduces a significance testing-enhanced conformal prediction (CP) framework to improve trustworthiness of large language models (LLMs) in multiple-choice question answering (MCQA). While LLMs have been increasingly deployed in disciplinary QA scenarios, hallucination and nonfactual generation substantially compromise response reliability. Although CP provides statistically rigorous marginal coverage guarantees for prediction sets, and significance testing offers established statistical rigor, their synergistic integration remains unexplored. To mitigate hallucination and factual inaccuracies, our framework integrates $p$-value computation with conformity scoring through self-consistency resampling of MCQA responses. This approach calculates option frequencies to address LLMs' black-box nature, subsequently constructing prediction sets via null hypothesis testing ($\\mathcal{H}_0$) with empirically derived $p$-values. Evaluations on MMLU and MMLU-Pro benchmarks using off-the-shelf LLMs demonstrate: (1) The enhanced CP achieves user-specified empirical miscoverage rates; (2) Test-set average prediction set size (APSS) decreases monotonically with increasing risk levels ($\\alpha$), validating APSS as an effective uncertainty metric. This work establishes a principled statistical framework for trustworthy LLM deployment in high-stakes QA applications.", "date": "2025-08-07", "categories": []} {"id": "2508.10284", "title": "Uncertainty-Aware Prediction of Parkinson's Disease Medication Needs: A Two-Stage Conformal Prediction Approach", "abstract": "Parkinson's Disease (PD) medication management presents unique challenges due to heterogeneous disease progression and treatment response. Neurologists must balance symptom control with optimal dopaminergic dosing based on functional disability while minimizing side effects. This balance is crucial as inadequate or abrupt changes can cause levodopa-induced dyskinesia, wearing off, and neuropsychiatric effects, significantly reducing quality of life. Current approaches rely on trial-and-error decisions without systematic predictive methods. Despite machine learning advances, clinical adoption remains limited due to reliance on point predictions that do not account for prediction uncertainty, undermining clinical trust and utility. Clinicians require not only predictions of future medication needs but also reliable confidence measures. Without quantified uncertainty, adjustments risk premature escalation to maximum doses or prolonged inadequate symptom control. We developed a conformal prediction framework anticipating medication needs up to two years in advance with reliable prediction intervals and statistical guarantees. Our approach addresses zero-inflation in PD inpatient data, where patients maintain stable medication regimens between visits. Using electronic health records from 631 inpatient admissions at University of Florida Health (2011-2021), our two-stage approach identifies patients likely to need medication changes, then predicts required levodopa equivalent daily dose adjustments. Our framework achieved marginal coverage while reducing prediction interval lengths compared to traditional approaches, providing precise predictions for short-term planning and wider ranges for long-term forecasting. By quantifying uncertainty, our approach enables evidence-based decisions about levodopa dosing, optimizing symptom control while minimizing side effects and improving life quality.", "date": "2025-08-14", "categories": []} {"id": "2406.01416", "title": "Adapting Prediction Sets to Distribution Shifts Without Labels", "abstract": "Recently there has been a surge of interest to deploy confidence set predictions rather than point predictions in machine learning. Unfortunately, the effectiveness of such prediction sets is frequently impaired by distribution shifts in practice, and the challenge is often compounded by the lack of ground truth labels at test time. Focusing on a standard set-valued prediction framework called conformal prediction (CP), this paper studies how to improve its practical performance using only unlabeled data from the shifted test domain. This is achieved by two new methods called ECP and EACP, whose main idea is to adjust the score function in CP according to its base model's own uncertainty evaluation. Through extensive experiments on a number of large-scale datasets and neural network architectures, we show that our methods provide consistent improvement over existing baselines and nearly match the performance of fully supervised methods.", "date": "2024-06-03", "categories": []} {"id": "2501.12749", "title": "Conformal Prediction of Classifiers with Many Classes based on Noisy Labels", "abstract": "Conformal Prediction (CP) controls the prediction uncertainty of classification systems by producing a small prediction set, ensuring a predetermined probability that the true class lies within this set. This is commonly done by defining a score, based on the model predictions, and setting a threshold on this score using a validation set. In this study, we address the problem of CP calibration when we only have access to a calibration set with noisy labels. We show how we can estimate the noise-free conformal threshold based on the noisy labeled data. We derive a finite sample coverage guarantee for uniform noise that remains effective even in tasks with a large number of classes. We dub our approach Noise-Aware Conformal Prediction (NACP). We illustrate the performance of the proposed results on several standard image classification datasets with a large number of classes.", "date": "2025-01-22", "categories": []} {"id": "2508.06885", "title": "Conformal Prediction and Trustworthy AI", "abstract": "Conformal predictors are machine learning algorithms developed in the 1990's by Gammerman, Vovk, and their research team, to provide set predictions with guaranteed confidence level. Over recent years, they have grown in popularity and have become a mainstream methodology for uncertainty quantification in the machine learning community. From its beginning, there was an understanding that they enable reliable machine learning with well-calibrated uncertainty quantification. This makes them extremely beneficial for developing trustworthy AI, a topic that has also risen in interest over the past few years, in both the AI community and society more widely. In this article, we review the potential for conformal prediction to contribute to trustworthy AI beyond its marginal validity property, addressing problems such as generalization risk and AI governance. Experiments and examples are also provided to demonstrate its use as a well-calibrated predictor and for bias identification and mitigation.", "date": "2025-08-09", "categories": []} {"id": "2508.05544", "title": "Conformal Sets in Multiple-Choice Question Answering under Black-Box Settings with Provable Coverage Guarantees", "abstract": "Large Language Models (LLMs) have shown remarkable progress in multiple-choice question answering (MCQA), but their inherent unreliability, such as hallucination and overconfidence, limits their application in high-risk domains. To address this, we propose a frequency-based uncertainty quantification method under black-box settings, leveraging conformal prediction (CP) to ensure provable coverage guarantees. Our approach involves multiple independent samplings of the model's output distribution for each input, with the most frequent sample serving as a reference to calculate predictive entropy (PE). Experimental evaluations across six LLMs and four datasets (MedMCQA, MedQA, MMLU, MMLU-Pro) demonstrate that frequency-based PE outperforms logit-based PE in distinguishing between correct and incorrect predictions, as measured by AUROC. Furthermore, the method effectively controls the empirical miscoverage rate under user-specified risk levels, validating that sampling frequency can serve as a viable substitute for logit-based probabilities in black-box scenarios. This work provides a distribution-free model-agnostic framework for reliable uncertainty quantification in MCQA with guaranteed coverage, enhancing the trustworthiness of LLMs in practical applications.", "date": "2025-08-07", "categories": []} {"id": "2505.17340", "title": "Conformal Predictive Distributions for Order Fulfillment Time Forecasting", "abstract": "Accurate estimation of order fulfillment time is critical for e-commerce logistics, yet traditional rule-based approaches often fail to capture the inherent uncertainties in delivery operations. This paper introduces a novel framework for distributional forecasting of order fulfillment time, leveraging Conformal Predictive Systems and Cross Venn-Abers Predictors -- model-agnostic techniques that provide rigorous coverage or validity guarantees. The proposed machine learning methods integrate granular spatiotemporal features, capturing fulfillment location and carrier performance dynamics to enhance predictive accuracy. Additionally, a cost-sensitive decision rule is developed to convert probabilistic forecasts into reliable point predictions. Experimental evaluation on a large-scale industrial dataset demonstrates that the proposed methods generate competitive distributional forecasts, while machine learning-based point predictions significantly outperform the existing rule-based system -- achieving up to 14% higher prediction accuracy and up to 75% improvement in identifying late deliveries.", "date": "2025-05-22", "categories": []} {"id": "2502.17264", "title": "Kandinsky Conformal Prediction: Beyond Class- and Covariate-Conditional Coverage", "abstract": "Conformal prediction is a powerful distribution-free framework for constructing prediction sets with coverage guarantees. Classical methods, such as split conformal prediction, provide marginal coverage, ensuring that the prediction set contains the label of a random test point with a target probability. However, these guarantees may not hold uniformly across different subpopulations, leading to disparities in coverage. Prior work has explored coverage guarantees conditioned on events related to the covariates and label of the test point. We present Kandinsky conformal prediction, a framework that significantly expands the scope of conditional coverage guarantees. In contrast to Mondrian conformal prediction, which restricts its coverage guarantees to disjoint groups -- reminiscent of the rigid, structured grids of Piet Mondrian's art -- our framework flexibly handles overlapping and fractional group memberships defined jointly on covariates and labels, reflecting the layered, intersecting forms in Wassily Kandinsky's compositions. Our algorithm unifies and extends existing methods, encompassing covariate-based group conditional, class conditional, and Mondrian conformal prediction as special cases, while achieving a minimax-optimal high-probability conditional coverage bound. Finally, we demonstrate the practicality of our approach through empirical evaluation on real-world datasets.", "date": "2025-02-24", "categories": []} {"id": "2507.20272", "title": "Approximating Full Conformal Prediction for Neural Network Regression with Gauss-Newton Influence", "abstract": "Uncertainty quantification is an important prerequisite for the deployment of deep learning models in safety-critical areas. Yet, this hinges on the uncertainty estimates being useful to the extent the prediction intervals are well-calibrated and sharp. In the absence of inherent uncertainty estimates (e.g. pretrained models predicting only point estimates), popular approaches that operate post-hoc include Laplace's method and split conformal prediction (split-CP). However, Laplace's method can be miscalibrated when the model is misspecified and split-CP requires sample splitting, and thus comes at the expense of statistical efficiency. In this work, we construct prediction intervals for neural network regressors post-hoc without held-out data. This is achieved by approximating the full conformal prediction method (full-CP). Whilst full-CP nominally requires retraining the model for every test point and candidate label, we propose to train just once and locally perturb model parameters using Gauss-Newton influence to approximate the effect of retraining. Coupled with linearization of the network, we express the absolute residual nonconformity score as a piecewise linear function of the candidate label allowing for an efficient procedure that avoids the exhaustive search over the output space. On standard regression benchmarks and bounding box localization, we show the resulting prediction intervals are locally-adaptive and often tighter than those of split-CP.", "date": "2025-07-27", "categories": []} {"id": "2507.16829", "title": "You Don't Bring Me Flowers: Mitigating Unwanted Recommendations Through Conformal Risk Control", "abstract": "Recommenders are significantly shaping online information consumption. While effective at personalizing content, these systems increasingly face criticism for propagating irrelevant, unwanted, and even harmful recommendations. Such content degrades user satisfaction and contributes to significant societal issues, including misinformation, radicalization, and erosion of user trust. Although platforms offer mechanisms to mitigate exposure to undesired content, these mechanisms are often insufficiently effective and slow to adapt to users' feedback. This paper introduces an intuitive, model-agnostic, and distribution-free method that uses conformal risk control to provably bound unwanted content in personalized recommendations by leveraging simple binary feedback on items. We also address a limitation of traditional conformal risk control approaches, i.e., the fact that the recommender can provide a smaller set of recommended items, by leveraging implicit feedback on consumed items to expand the recommendation set while ensuring robust risk mitigation. Our experimental evaluation on data coming from a popular online video-sharing platform demonstrates that our approach ensures an effective and controllable reduction of unwanted recommendations with minimal effort. The source code is available here: https://github.com/geektoni/mitigating-harm-recsys.", "date": "2025-07-09", "categories": []} {"id": "2507.15741", "title": "Conformal and kNN Predictive Uncertainty Quantification Algorithms in Metric Spaces", "abstract": "This paper introduces a framework for uncertainty quantification in regression models defined in metric spaces. Leveraging a newly defined notion of homoscedasticity, we develop a conformal prediction algorithm that offers finite-sample coverage guarantees and fast convergence rates of the oracle estimator. In heteroscedastic settings, we forgo these non-asymptotic guarantees to gain statistical efficiency, proposing a local $k$--nearest--neighbor method without conformal calibration that is adaptive to the geometry of each particular nonlinear space. Both procedures work with any regression algorithm and are scalable to large data sets, allowing practitioners to plug in their preferred models and incorporate domain expertise. We prove consistency for the proposed estimators under minimal conditions. Finally, we demonstrate the practical utility of our approach in personalized--medicine applications involving random response objects such as probability distributions and graph Laplacians.", "date": "2025-07-21", "categories": []} {"id": "2507.15825", "title": "ACS: An interactive framework for conformal selection", "abstract": "This paper presents adaptive conformal selection (ACS), an interactive framework for model-free selection with guaranteed error control. Building on conformal selection (Jin and Cand\\`es, 2023b), ACS generalizes the approach to support human-in-the-loop adaptive data analysis. Under the ACS framework, we can partially reuse the data to boost the selection power, make decisions on the fly while exploring the data, and incorporate new information or preferences as they arise. The key to ACS is a carefully designed principle that controls the information available for decision making, allowing the data analyst to explore the data adaptively while maintaining rigorous control of the false discovery rate (FDR). Based on the ACS framework, we provide concrete selection algorithms for various goals, including model update/selection, diversified selection, and incorporating newly available labeled data. The effectiveness of ACS is demonstrated through extensive numerical simulations and real-data applications in large language model (LLM) deployment and drug discovery.", "date": "2025-07-21", "categories": []} {"id": "2507.13835", "title": "Conformal Data Contamination Tests for Trading or Sharing of Data", "abstract": "The amount of quality data in many machine learning tasks is limited to what is available locally to data owners. The set of quality data can be expanded through trading or sharing with external data agents. However, data buyers need quality guarantees before purchasing, as external data may be contaminated or irrelevant to their specific learning task. Previous works primarily rely on distributional assumptions about data from different agents, relegating quality checks to post-hoc steps involving costly data valuation procedures. We propose a distribution-free, contamination-aware data-sharing framework that identifies external data agents whose data is most valuable for model personalization. To achieve this, we introduce novel two-sample testing procedures, grounded in rigorous theoretical foundations for conformal outlier detection, to determine whether an agent's data exceeds a contamination threshold. The proposed tests, termed conformal data contamination tests, remain valid under arbitrary contamination levels while enabling false discovery rate control via the Benjamini-Hochberg procedure. Empirical evaluations across diverse collaborative learning scenarios demonstrate the robustness and effectiveness of our approach. Overall, the conformal data contamination test distinguishes itself as a generic procedure for aggregating data with statistically rigorous quality guarantees.", "date": "2025-07-18", "categories": []} {"id": "2507.14023", "title": "Conformalized Regression for Continuous Bounded Outcomes", "abstract": "Regression problems with bounded continuous outcomes frequently arise in real-world statistical and machine learning applications, such as the analysis of rates and proportions. A central challenge in this setting is predicting a response associated with a new covariate value. Most of the existing statistical and machine learning literature has focused either on point prediction of bounded outcomes or on interval prediction based on asymptotic approximations. We develop conformal prediction intervals for bounded outcomes based on transformation models and beta regression. We introduce tailored non-conformity measures based on residuals that are aligned with the underlying models, and account for the inherent heteroscedasticity in regression settings with bounded outcomes. We present a theoretical result on asymptotic marginal and conditional validity in the context of full conformal prediction, which remains valid under model misspecification. For split conformal prediction, we provide an empirical coverage analysis based on a comprehensive simulation study. The simulation study demonstrates that both methods provide valid finite-sample predictive coverage, including settings with model misspecification. Finally, we demonstrate the practical performance of the proposed conformal prediction intervals on real data and compare them with bootstrap-based alternatives.", "date": "2025-07-18", "categories": []} {"id": "2310.08209", "title": "Conformal inference for regression on Riemannian Manifolds", "abstract": "Regression on manifolds, and, more broadly, statistics on manifolds, has garnered significant importance in recent years due to the vast number of applications for non Euclidean data. Circular data is a classic example, but so is data in the space of covariance matrices, data on the Grassmannian manifold obtained as a result of principal component analysis, among many others. In this work we investigate prediction sets for regression scenarios when the response variable, denoted by $Y$, resides in a manifold, and the covariable, denoted by $X$, lies in an Euclidean space. This extends the concepts delineated in \\cite{waser14} to this novel context. Aligning with traditional principles in conformal inference, these prediction sets are distribution-free, indicating that no specific assumptions are imposed on the joint distribution of $(X,Y)$, and they maintain a non-parametric character. We prove the asymptotic almost sure convergence of the empirical version of these regions on the manifold to their population counterparts. The efficiency of this method is shown through a comprehensive simulation study and an analysis involving real-world data.", "date": "2023-10-12", "categories": []} {"id": "2507.12497", "title": "Differentially Private Conformal Prediction via Quantile Binary Search", "abstract": "Most Differentially Private (DP) approaches focus on limiting privacy leakage from learners based on the data that they are trained on, there are fewer approaches that consider leakage when procedures involve a calibration dataset which is common in uncertainty quantification methods such as Conformal Prediction (CP). Since there is a limited amount of approaches in this direction, in this work we deliver a general DP approach for CP that we call Private Conformity via Quantile Search (P-COQS). The proposed approach adapts an existing randomized binary search algorithm for computing DP quantiles in the calibration phase of CP thereby guaranteeing privacy of the consequent prediction sets. This however comes at a price of slightly under-covering with respect to the desired $(1 - \\alpha)$-level when using finite-sample calibration sets (although broad empirical results show that the P-COQS generally targets the required level in the considered cases). Confirming properties of the adapted algorithm and quantifying the approximate coverage guarantees of the consequent CP, we conduct extensive experiments to examine the effects of privacy noise, sample size and significance level on the performance of our approach compared to existing alternatives. In addition, we empirically evaluate our approach on several benchmark datasets, including CIFAR-10, ImageNet and CoronaHack. Our results suggest that the proposed method is robust to privacy noise and performs favorably with respect to the current DP alternative in terms of empirical coverage, efficiency, and informativeness. Specifically, the results indicate that P-COQS produces smaller conformal prediction sets while simultaneously targeting the desired coverage and privacy guarantees in all these experimental settings.", "date": "2025-07-15", "categories": []} {"id": "2507.11574", "title": "Distribution-Free Uncertainty-Aware Virtual Sensing via Conformalized Neural Operators", "abstract": "Robust uncertainty quantification (UQ) remains a critical barrier to the safe deployment of deep learning in real-time virtual sensing, particularly in high-stakes domains where sparse, noisy, or non-collocated sensor data are the norm. We introduce the Conformalized Monte Carlo Operator (CMCO), a framework that transforms neural operator-based virtual sensing with calibrated, distribution-free prediction intervals. By unifying Monte Carlo dropout with split conformal prediction in a single DeepONet architecture, CMCO achieves spatially resolved uncertainty estimates without retraining, ensembling, or custom loss design. Our method addresses a longstanding challenge: how to endow operator learning with efficient and reliable UQ across heterogeneous domains. Through rigorous evaluation on three distinct applications: turbulent flow, elastoplastic deformation, and global cosmic radiation dose estimation-CMCO consistently attains near-nominal empirical coverage, even in settings with strong spatial gradients and proxy-based sensing. This breakthrough offers a general-purpose, plug-and-play UQ solution for neural operators, unlocking real-time, trustworthy inference in digital twins, sensor fusion, and safety-critical monitoring. By bridging theory and deployment with minimal computational overhead, CMCO establishes a new foundation for scalable, generalizable, and uncertainty-aware scientific machine learning.", "date": "2025-07-15", "categories": []} {"id": "2507.11739", "title": "Sparse Identification of Nonlinear Dynamics with Conformal Prediction", "abstract": "The Sparse Identification of Nonlinear Dynamics (SINDy) is a method for discovering nonlinear dynamical system models from data. Quantifying uncertainty in SINDy models is essential for assessing their reliability, particularly in safety-critical applications. While various uncertainty quantification methods exist for SINDy, including Bayesian and ensemble approaches, this work explores the integration of Conformal Prediction, a framework that can provide valid prediction intervals with coverage guarantees based on minimal assumptions like data exchangeability. We introduce three applications of conformal prediction with Ensemble-SINDy (E-SINDy): (1) quantifying uncertainty in time series prediction, (2) model selection based on library feature importance, and (3) quantifying the uncertainty of identified model coefficients using feature conformal prediction. We demonstrate the three applications on stochastic predator-prey dynamics and several chaotic dynamical systems. We show that conformal prediction methods integrated with E-SINDy can reliably achieve desired target coverage for time series forecasting, effectively quantify feature importance, and produce more robust uncertainty intervals for model coefficients, even under non-Gaussian noise, compared to standard E-SINDy coefficient estimates.", "date": "2025-07-15", "categories": []} {"id": "2501.00555", "title": "Prune 'n Predict: Optimizing LLM Decision-making with Conformal Prediction", "abstract": "Large language models (LLMs) are empowering decision-making in several applications, including tool or API usage and answering multiple-choice questions (MCQs). However, incorrect outputs pose significant risks in high-stakes domains like healthcare and finance. To quantify LLM uncertainty and thereby mitigate these risks, recent works employ conformal prediction (CP), a model- and distribution-agnostic framework that uses LLM outputs to generate a \\emph{prediction set} containing the true answer with high probability. Leveraging CP, we propose \\emph{conformal revision of questions} (CROQ), which revises the question by narrowing down the available choices to those in the prediction set and asking the LLM the revised question. We expect LLMs to be more accurate on revised questions with fewer choices. Furthermore, we expect CROQ to be effective when the prediction sets from CP are small. Commonly used logit scores often lead to large sets, diminishing CROQ's effectiveness. To overcome this, we propose CP-OPT, an optimization framework to learn scores that minimize set sizes while maintaining coverage. Our extensive experiments on MMLU, ToolAlpaca, and TruthfulQA datasets with multiple LLMs show that CROQ improves accuracy over the standard inference, with more pronounced gains when paired with CP-OPT.", "date": "2024-12-31", "categories": []} {"id": "2507.08858", "title": "Foundation models for time series forecasting: Application in conformal prediction", "abstract": "The zero-shot capabilities of foundation models (FMs) for time series forecasting offer promising potentials in conformal prediction, as most of the available data can be allocated to calibration. This study compares the performance of Time Series Foundation Models (TSFMs) with traditional methods, including statistical models and gradient boosting, within a conformal prediction setting. Our findings highlight two key advantages of TSFMs. First, when the volume of data is limited, TSFMs provide more reliable conformalized prediction intervals than classic models, thanks to their superior predictive accuracy. Second, the calibration process is more stable because more data are used for calibration. Morever, the fewer data available, the more pronounced these benefits become, as classic models require a substantial amount of data for effective training. These results underscore the potential of foundation models in improving conformal prediction reliability in time series applications, particularly in data-constrained cases. All the code to reproduce the experiments is available.", "date": "2025-07-09", "categories": []} {"id": "2507.09678", "title": "Conformal Prediction for Privacy-Preserving Machine Learning", "abstract": "We investigate the integration of Conformal Prediction (CP) with supervised learning on deterministically encrypted data, aiming to bridge the gap between rigorous uncertainty quantification and privacy-preserving machine learning. Using AES-encrypted variants of the MNIST dataset, we demonstrate that CP methods remain effective even when applied directly in the encrypted domain, owing to the preservation of data exchangeability under fixed-key encryption. We test traditional $p$-value-based against $e$-value-based conformal predictors. Our empirical evaluation reveals that models trained on deterministically encrypted data retain the ability to extract meaningful structure, achieving 36.88\\% test accuracy -- significantly above random guessing (9.56\\%) observed with per-instance encryption. Moreover, $e$-value-based CP achieves predictive set coverage of over 60\\% with 4.3 loss-threshold calibration, correctly capturing the true label in 4888 out of 5000 test cases. In contrast, the $p$-value-based CP yields smaller predictive sets but with reduced coverage accuracy. These findings highlight both the promise and limitations of CP in encrypted data settings and underscore critical trade-offs between prediction set compactness and reliability. %Our work sets a foundation for principled uncertainty quantification in secure, privacy-aware learning systems.", "date": "2025-07-13", "categories": []} {"id": "2502.03023", "title": "Parametric Scaling Law of Tuning Bias in Conformal Prediction", "abstract": "Conformal prediction is a popular framework of uncertainty quantification that constructs prediction sets with coverage guarantees. To uphold the exchangeability assumption, many conformal prediction methods necessitate an additional holdout set for parameter tuning. Yet, the impact of violating this principle on coverage remains underexplored, making it ambiguous in practical applications. In this work, we empirically find that the tuning bias - the coverage gap introduced by leveraging the same dataset for tuning and calibration, is negligible for simple parameter tuning in many conformal prediction methods. In particular, we observe the scaling law of the tuning bias: this bias increases with parameter space complexity and decreases with calibration set size. Formally, we establish a theoretical framework to quantify the tuning bias and provide rigorous proof for the scaling law of the tuning bias by deriving its upper bound. In the end, we discuss how to reduce the tuning bias, guided by the theories we developed.", "date": "2025-02-05", "categories": []} {"id": "2507.06921", "title": "Distribution-free inference for LightGBM and GLM with Tweedie loss", "abstract": "Prediction uncertainty quantification is a key research topic in recent years scientific and business problems. In insurance industries (\\cite{parodi2023pricing}), assessing the range of possible claim costs for individual drivers improves premium pricing accuracy. It also enables insurers to manage risk more effectively by accounting for uncertainty in accident likelihood and severity. In the presence of covariates, a variety of regression-type models are often used for modeling insurance claims, ranging from relatively simple generalized linear models (GLMs) to regularized GLMs to gradient boosting models (GBMs). Conformal predictive inference has arisen as a popular distribution-free approach for quantifying predictive uncertainty under relatively weak assumptions of exchangeability, and has been well studied under the classic linear regression setting. In this work, we propose new non-conformity measures for GLMs and GBMs with GLM-type loss. Using regularized Tweedie GLM regression and LightGBM with Tweedie loss, we demonstrate conformal prediction performance with these non-conformity measures in insurance claims data. Our simulation results favor the use of locally weighted Pearson residuals for LightGBM over other methods considered, as the resulting intervals maintained the nominal coverage with the smallest average width.", "date": "2025-07-09", "categories": []} {"id": "2503.02803", "title": "Inductive randomness predictors: beyond conformal", "abstract": "This paper introduces inductive randomness predictors, which form a proper superset of inductive conformal predictors but have the same principal property of validity under the assumption of randomness (i.e., of IID data). It turns out that every non-trivial inductive conformal predictor is strictly dominated by an inductive randomness predictor, although the improvement is not great, at most a factor of $\\mathrm{e}\\approx2.72$ in the case of e-prediction. The dominating inductive randomness predictors are more complicated and more difficult to compute; besides, an improvement by a factor of $\\mathrm{e}$ is rare. Therefore, this paper does not suggest replacing inductive conformal predictors by inductive randomness predictors and only calls for a more detailed study of the latter.", "date": "2025-03-04", "categories": []} {"id": "2503.10345", "title": "Mirror Online Conformal Prediction with Intermittent Feedback", "abstract": "Online conformal prediction enables the runtime calibration of a pre-trained artificial intelligence model using feedback on its performance. Calibration is achieved through set predictions that are updated via online rules so as to ensure long-term coverage guarantees. While recent research has demonstrated the benefits of incorporating prior knowledge into the calibration process, this has come at the cost of replacing coverage guarantees with less tangible regret guarantees based on the quantile loss. This work introduces intermittent mirror online conformal prediction (IM-OCP), a novel runtime calibration framework that integrates prior knowledge, operates under potentially intermittent feedback, and features minimal memory complexity. IM-OCP guarantees long-term coverage and sub-linear regret, both of which hold deterministically for any given data sequence and in expectation with respect to the intermittent feedback.", "date": "2025-03-13", "categories": []} {"id": "2507.02151", "title": "Non-exchangeable Conformal Prediction for Temporal Graph Neural Networks", "abstract": "Conformal prediction for graph neural networks (GNNs) offers a promising framework for quantifying uncertainty, enhancing GNN reliability in high-stakes applications. However, existing methods predominantly focus on static graphs, neglecting the evolving nature of real-world graphs. Temporal dependencies in graph structure, node attributes, and ground truth labels violate the fundamental exchangeability assumption of standard conformal prediction methods, limiting their applicability. To address these challenges, in this paper, we introduce NCPNET, a novel end-to-end conformal prediction framework tailored for temporal graphs. Our approach extends conformal prediction to dynamic settings, mitigating statistical coverage violations induced by temporal dependencies. To achieve this, we propose a diffusion-based non-conformity score that captures both topological and temporal uncertainties within evolving networks. Additionally, we develop an efficiency-aware optimization algorithm that improves the conformal prediction process, enhancing computational efficiency and reducing coverage violations. Extensive experiments on diverse real-world temporal graphs, including WIKI, REDDIT, DBLP, and IBM Anti-Money Laundering dataset, demonstrate NCPNET's capability to ensure guaranteed coverage in temporal graphs, achieving up to a 31% reduction in prediction set size on the WIKI dataset, significantly improving efficiency compared to state-of-the-art methods. Our data and code are available at https://github.com/ODYSSEYWT/NCPNET.", "date": "2025-07-02", "categories": []} {"id": "2409.15548", "title": "Beyond Conformal Predictors: Adaptive Conformal Inference with Confidence Predictors", "abstract": "Adaptive Conformal Inference (ACI) provides finite-sample coverage guarantees, enhancing the prediction reliability under non-exchangeability. This study demonstrates that these desirable properties of ACI do not require the use of Conformal Predictors (CP). We show that the guarantees hold for the broader class of confidence predictors, defined by the requirement of producing nested prediction sets, a property we argue is essential for meaningful confidence statements. We empirically investigate the performance of Non-Conformal Confidence Predictors (NCCP) against CP when used with ACI on non-exchangeable data. In online settings, the NCCP offers significant computational advantages while maintaining a comparable predictive efficiency. In batch settings, inductive NCCP (INCCP) can outperform inductive CP (ICP) by utilising the full training dataset without requiring a separate calibration set, leading to improved efficiency, particularly when the data are limited. Although these initial results highlight NCCP as a theoretically sound and practically effective alternative to CP for uncertainty quantification with ACI in non-exchangeable scenarios, further empirical studies are warranted across diverse datasets and predictors.", "date": "2024-09-23", "categories": []} {"id": "2502.13030", "title": "Conformal Inference under High-Dimensional Covariate Shifts via Likelihood-Ratio Regularization", "abstract": "We consider the problem of conformal prediction under covariate shift. Given labeled data from a source domain and unlabeled data from a covariate shifted target domain, we seek to construct prediction sets with valid marginal coverage in the target domain. Most existing methods require estimating the unknown likelihood ratio function, which can be prohibitive for high-dimensional data such as images. To address this challenge, we introduce the likelihood ratio regularized quantile regression (LR-QR) algorithm, which combines the pinball loss with a novel choice of regularization in order to construct a threshold function without directly estimating the unknown likelihood ratio. We show that the LR-QR method has coverage at the desired level in the target domain, up to a small error term that we can control. Our proofs draw on a novel analysis of coverage via stability bounds from learning theory. Our experiments demonstrate that the LR-QR algorithm outperforms existing methods on high-dimensional prediction tasks, including a regression task for the Communities and Crime dataset, an image classification task from the WILDS repository, and an LLM question-answering task on the MMLU benchmark.", "date": "2025-02-18", "categories": []} {"id": "2504.14154", "title": "SConU: Selective Conformal Uncertainty in Large Language Models", "abstract": "As large language models are increasingly utilized in real-world applications, guarantees of task-specific metrics are essential for their reliable deployment. Previous studies have introduced various criteria of conformal uncertainty grounded in split conformal prediction, which offer user-specified correctness coverage. However, existing frameworks often fail to identify uncertainty data outliers that violate the exchangeability assumption, leading to unbounded miscoverage rates and unactionable prediction sets. In this paper, we propose a novel approach termed Selective Conformal Uncertainty (SConU), which, for the first time, implements significance tests, by developing two conformal p-values that are instrumental in determining whether a given sample deviates from the uncertainty distribution of the calibration set at a specific manageable risk level. Our approach not only facilitates rigorous management of miscoverage rates across both single-domain and interdisciplinary contexts, but also enhances the efficiency of predictions. Furthermore, we comprehensively analyze the components of the conformal procedures, aiming to approximate conditional coverage, particularly in high-stakes question-answering tasks.", "date": "2025-04-19", "categories": []} {"id": "2506.21802", "title": "Classification with Reject Option: Distribution-free Error Guarantees via Conformal Prediction", "abstract": "Machine learning (ML) models always make a prediction, even when they are likely to be wrong. This causes problems in practical applications, as we do not know if we should trust a prediction. ML with reject option addresses this issue by abstaining from making a prediction if it is likely to be incorrect. In this work, we formalise the approach to ML with reject option in binary classification, deriving theoretical guarantees on the resulting error rate. This is achieved through conformal prediction (CP), which produce prediction sets with distribution-free validity guarantees. In binary classification, CP can output prediction sets containing exactly one, two or no labels. By accepting only the singleton predictions, we turn CP into a binary classifier with reject option.\n Here, CP is formally put in the framework of predicting with reject option. We state and prove the resulting error rate, and give finite sample estimates. Numerical examples provide illustrations of derived error rate through several different conformal prediction settings, ranging from full conformal prediction to offline batch inductive conformal prediction. The former has a direct link to sharp validity guarantees, whereas the latter is more fuzzy in terms of validity guarantees but can be used in practice. Error-reject curves illustrate the trade-off between error rate and reject rate, and can serve to aid a user to set an acceptable error rate or reject rate in practice.", "date": "2025-06-26", "categories": []} {"id": "2506.20173", "title": "Valid Selection among Conformal Sets", "abstract": "Conformal prediction offers a distribution-free framework for constructing prediction sets with coverage guarantees. In practice, multiple valid conformal prediction sets may be available, arising from different models or methodologies. However, selecting the most desirable set, such as the smallest, can invalidate the coverage guarantees. To address this challenge, we propose a stability-based approach that ensures coverage for the selected prediction set. We extend our results to the online conformal setting, propose several refinements in settings where additional structure is available, and demonstrate its effectiveness through experiments.", "date": "2025-06-25", "categories": []} {"id": "2506.19689", "title": "When Can We Reuse a Calibration Set for Multiple Conformal Predictions?", "abstract": "Reliable uncertainty quantification is crucial for the trustworthiness of machine learning applications. Inductive Conformal Prediction (ICP) offers a distribution-free framework for generating prediction sets or intervals with user-specified confidence. However, standard ICP guarantees are marginal and typically require a fresh calibration set for each new prediction to maintain their validity. This paper addresses this practical limitation by demonstrating how e-conformal prediction, in conjunction with Hoeffding's inequality, can enable the repeated use of a single calibration set with a high probability of preserving the desired coverage. Through a case study on the CIFAR-10 dataset, we train a deep neural network and utilise a calibration set to estimate a Hoeffding correction. This correction allows us to apply a modified Markov's inequality, leading to the construction of prediction sets with quantifiable confidence. Our results illustrate the feasibility of maintaining provable performance in conformal prediction while enhancing its practicality by reducing the need for repeated calibration. The code for this work is publicly available.", "date": "2025-06-24", "categories": []} {"id": "2506.17503", "title": "Trustworthy Few-Shot Transfer of Medical VLMs through Split Conformal Prediction", "abstract": "Medical vision-language models (VLMs) have demonstrated unprecedented transfer capabilities and are being increasingly adopted for data-efficient image classification. Despite its growing popularity, its reliability aspect remains largely unexplored. This work explores the split conformal prediction (SCP) framework to provide trustworthiness guarantees when transferring such models based on a small labeled calibration set. Despite its potential, the generalist nature of the VLMs' pre-training could negatively affect the properties of the predicted conformal sets for specific tasks. While common practice in transfer learning for discriminative purposes involves an adaptation stage, we observe that deploying such a solution for conformal purposes is suboptimal since adapting the model using the available calibration data breaks the rigid exchangeability assumptions for test data in SCP. To address this issue, we propose transductive split conformal adaptation (SCA-T), a novel pipeline for transfer learning on conformal scenarios, which performs an unsupervised transductive adaptation jointly on calibration and test data. We present comprehensive experiments utilizing medical VLMs across various image modalities, transfer tasks, and non-conformity scores. Our framework offers consistent gains in efficiency and conditional coverage compared to SCP, maintaining the same empirical guarantees.", "date": "2025-06-20", "categories": []} {"id": "2506.18162", "title": "Pitfalls of Conformal Predictions for Medical Image Classification", "abstract": "Reliable uncertainty estimation is one of the major challenges for medical classification tasks. While many approaches have been proposed, recently the statistical framework of conformal predictions has gained a lot of attention, due to its ability to provide provable calibration guarantees. Nonetheless, the application of conformal predictions in safety-critical areas such as medicine comes with pitfalls, limitations and assumptions that practitioners need to be aware of. We demonstrate through examples from dermatology and histopathology that conformal predictions are unreliable under distributional shifts in input and label variables. Additionally, conformal predictions should not be used for selecting predictions to improve accuracy and are not reliable for subsets of the data, such as individual classes or patient attributes. Moreover, in classification settings with a small number of classes, which are common in medical image classification tasks, conformal predictions have limited practical value.", "date": "2025-06-22", "categories": []} {"id": "2501.18426", "title": "Guaranteed prediction sets for functional surrogate models", "abstract": "We propose a method for obtaining statistically guaranteed prediction sets for functional machine learning methods: surrogate models which map between function spaces, motivated by the need to build reliable PDE emulators. The method constructs nested prediction sets on a low-dimensional representation (an SVD) of the surrogate model's error, and then maps these sets to the prediction space using set-propagation techniques. This results in prediction sets for functional surrogate models with conformal prediction coverage guarantees. We use zonotopes as basis of the set construction, which allow an exact linear propagation and are closed under Cartesian products, making them well-suited to this high-dimensional problem. The method is model agnostic and can thus be applied to complex Sci-ML models, including Neural Operators, but also in simpler settings. We also introduce a technique to capture the truncation error of the SVD, preserving the guarantees of the method.", "date": "2025-01-30", "categories": []} {"id": "2506.16189", "title": "CP$^2$: Leveraging Geometry for Conformal Prediction via Canonicalization", "abstract": "We study the problem of conformal prediction (CP) under geometric data shifts, where data samples are susceptible to transformations such as rotations or flips. While CP endows prediction models with post-hoc uncertainty quantification and formal coverage guarantees, their practicality breaks under distribution shifts that deteriorate model performance. To address this issue, we propose integrating geometric information--such as geometric pose--into the conformal procedure to reinstate its guarantees and ensure robustness under geometric shifts. In particular, we explore recent advancements on pose canonicalization as a suitable information extractor for this purpose. Evaluating the combined approach across discrete and continuous shifts and against equivariant and augmentation-based baselines, we find that integrating geometric information with CP yields a principled way to address geometric shifts while maintaining broad applicability to black-box predictors.", "date": "2025-06-19", "categories": []} {"id": "2503.10512", "title": "Conformal Prediction Sets for Deep Generative Models via Reduction to Conformal Regression", "abstract": "We consider the problem of generating valid and small prediction sets by sampling outputs (e.g., software code and natural language text) from a black-box deep generative model for a given input (e.g., textual prompt). The validity of a prediction set is determined by a user-defined binary admissibility function depending on the target application. For example, requiring at least one program in the set to pass all test cases in code generation application. To address this problem, we develop a simple and effective conformal inference algorithm referred to as Generative Prediction Sets (GPS). Given a set of calibration examples and black-box access to a deep generative model, GPS can generate prediction sets with provable guarantees. The key insight behind GPS is to exploit the inherent structure within the distribution over the minimum number of samples needed to obtain an admissible output to develop a simple conformal regression approach over the minimum number of samples. Experiments on multiple datasets for code and math word problems using different large language models demonstrate the efficacy of GPS over state-of-the-art methods.", "date": "2025-03-13", "categories": []} {"id": "2208.02814", "title": "Conformal Risk Control", "abstract": "We extend conformal prediction to control the expected value of any monotone loss function. The algorithm generalizes split conformal prediction together with its coverage guarantee. Like conformal prediction, the conformal risk control procedure is tight up to an $\\mathcal{O}(1/n)$ factor. We also introduce extensions of the idea to distribution shift, quantile risk control, multiple and adversarial risk control, and expectations of U-statistics. Worked examples from computer vision and natural language processing demonstrate the usage of our algorithm to bound the false negative rate, graph distance, and token-level F1-score.", "date": "2022-08-04", "categories": []} {"id": "2502.04807", "title": "Robust Conformal Outlier Detection under Contaminated Reference Data", "abstract": "Conformal prediction is a flexible framework for calibrating machine learning predictions, providing distribution-free statistical guarantees. In outlier detection, this calibration relies on a reference set of labeled inlier data to control the type-I error rate. However, obtaining a perfectly labeled inlier reference set is often unrealistic, and a more practical scenario involves access to a contaminated reference set containing a small fraction of outliers. This paper analyzes the impact of such contamination on the validity of conformal methods. We prove that under realistic, non-adversarial settings, calibration on contaminated data yields conservative type-I error control, shedding light on the inherent robustness of conformal methods. This conservativeness, however, typically results in a loss of power. To alleviate this limitation, we propose a novel, active data-cleaning framework that leverages a limited labeling budget and an outlier detection model to selectively annotate data points in the contaminated reference set that are suspected as outliers. By removing only the annotated outliers in this ``suspicious'' subset, we can effectively enhance power while mitigating the risk of inflating the type-I error rate, as supported by our theoretical analysis. Experiments on real datasets validate the conservative behavior of conformal methods under contamination and show that the proposed data-cleaning strategy improves power without sacrificing validity.", "date": "2025-02-07", "categories": []} {"id": "2505.13432", "title": "Synthetic-Powered Predictive Inference", "abstract": "Conformal prediction is a framework for predictive inference with a distribution-free, finite-sample guarantee. However, it tends to provide uninformative prediction sets when calibration data are scarce. This paper introduces Synthetic-powered predictive inference (SPI), a novel framework that incorporates synthetic data -- e.g., from a generative model -- to improve sample efficiency. At the core of our method is a score transporter: an empirical quantile mapping that aligns nonconformity scores from trusted, real data with those from synthetic data. By carefully integrating the score transporter into the calibration process, SPI provably achieves finite-sample coverage guarantees without making any assumptions about the real and synthetic data distributions. When the score distributions are well aligned, SPI yields substantially tighter and more informative prediction sets than standard conformal prediction. Experiments on image classification -- augmenting data with synthetic diffusion-model generated images -- and on tabular regression demonstrate notable improvements in predictive efficiency in data-scarce settings.", "date": "2025-05-19", "categories": []} {"id": "2502.13228", "title": "Conformal Prediction as Bayesian Quadrature", "abstract": "As machine learning-based prediction systems are increasingly used in high-stakes situations, it is important to understand how such predictive models will perform upon deployment. Distribution-free uncertainty quantification techniques such as conformal prediction provide guarantees about the loss black-box models will incur even when the details of the models are hidden. However, such methods are based on frequentist probability, which unduly limits their applicability. We revisit the central aspects of conformal prediction from a Bayesian perspective and thereby illuminate the shortcomings of frequentist guarantees. We propose a practical alternative based on Bayesian quadrature that provides interpretable guarantees and offers a richer representation of the likely range of losses to be observed at test time.", "date": "2025-02-18", "categories": []} {"id": "2502.06995", "title": "Epistemic Uncertainty in Conformal Scores: A Unified Approach", "abstract": "Conformal prediction methods create prediction bands with distribution-free guarantees but do not explicitly capture epistemic uncertainty, which can lead to overconfident predictions in data-sparse regions. Although recent conformal scores have been developed to address this limitation, they are typically designed for specific tasks, such as regression or quantile regression. Moreover, they rely on particular modeling choices for epistemic uncertainty, restricting their applicability. We introduce $\\texttt{EPICSCORE}$, a model-agnostic approach that enhances any conformal score by explicitly integrating epistemic uncertainty. Leveraging Bayesian techniques such as Gaussian Processes, Monte Carlo Dropout, or Bayesian Additive Regression Trees, $\\texttt{EPICSCORE}$ adaptively expands predictive intervals in regions with limited data while maintaining compact intervals where data is abundant. As with any conformal method, it preserves finite-sample marginal coverage. Additionally, it also achieves asymptotic conditional coverage. Experiments demonstrate its good performance compared to existing methods. Designed for compatibility with any Bayesian model, but equipped with distribution-free guarantees, $\\texttt{EPICSCORE}$ provides a general-purpose framework for uncertainty quantification in prediction problems.", "date": "2025-02-10", "categories": []} {"id": "2506.05434", "title": "Efficient Robust Conformal Prediction via Lipschitz-Bounded Networks", "abstract": "Conformal Prediction (CP) has proven to be an effective post-hoc method for improving the trustworthiness of neural networks by providing prediction sets with finite-sample guarantees. However, under adversarial attacks, classical conformal guarantees do not hold anymore: this problem is addressed in the field of Robust Conformal Prediction. Several methods have been proposed to provide robust CP sets with guarantees under adversarial perturbations, but, for large scale problems, these sets are either too large or the methods are too computationally demanding to be deployed in real life scenarios. In this work, we propose a new method that leverages Lipschitz-bounded networks to precisely and efficiently estimate robust CP sets. When combined with a 1-Lipschitz robust network, we demonstrate that our lip-rcp method outperforms state-of-the-art results in both the size of the robust CP sets and computational efficiency in medium and large-scale scenarios such as ImageNet. Taking a different angle, we also study vanilla CP under attack, and derive new worst-case coverage bounds of vanilla CP sets, which are valid simultaneously for all adversarial attack levels. Our lip-rcp method makes this second approach as efficient as vanilla CP while also allowing robustness guarantees.", "date": "2025-06-05", "categories": []} {"id": "2506.08325", "title": "Model-Free Kernel Conformal Depth Measures Algorithm for Uncertainty Quantification in Regression Models in Separable Hilbert Spaces", "abstract": "Depth measures are powerful tools for defining level sets in emerging, non--standard, and complex random objects such as high-dimensional multivariate data, functional data, and random graphs. Despite their favorable theoretical properties, the integration of depth measures into regression modeling to provide prediction regions remains a largely underexplored area of research. To address this gap, we propose a novel, model-free uncertainty quantification algorithm based on conditional depth measures--specifically, conditional kernel mean embeddings and an integrated depth measure. These new algorithms can be used to define prediction and tolerance regions when predictors and responses are defined in separable Hilbert spaces. The use of kernel mean embeddings ensures faster convergence rates in prediction region estimation. To enhance the practical utility of the algorithms with finite samples, we also introduce a conformal prediction variant that provides marginal, non-asymptotic guarantees for the derived prediction regions. Additionally, we establish both conditional and unconditional consistency results, as well as fast convergence rates in certain homoscedastic settings. We evaluate the finite--sample performance of our model in extensive simulation studies involving various types of functional data and traditional Euclidean scenarios. Finally, we demonstrate the practical relevance of our approach through a digital health application related to physical activity, aiming to provide personalized recommendations", "date": "2025-06-10", "categories": []} {"id": "2501.11689", "title": "Randomness, exchangeability, and conformal prediction", "abstract": "This paper argues for a wider use of the functional theory of randomness, a modification of the algorithmic theory of randomness getting rid of unspecified additive constants. Both theories are useful for understanding relationships between the assumptions of IID data and data exchangeability. While the assumption of IID data is standard in machine learning, conformal prediction relies on data exchangeability. Nouretdinov, V'yugin, and Gammerman showed, using the language of the algorithmic theory of randomness, that conformal prediction is a universal method under the assumption of IID data. In this paper (written for the Alex Gammerman Festschrift) I will selectively review connections between exchangeability and the property of being IID, early history of conformal prediction, my encounters and collaboration with Alex and other interesting people, and a translation of Nouretdinov et al.'s results into the language of the functional theory of randomness, which moves it closer to practice. Namely, the translation says that every confidence predictor that is valid for IID data can be transformed to a conformal predictor without losing much in predictive efficiency.", "date": "2025-01-20", "categories": []} {"id": "2502.19254", "title": "Universality of conformal prediction under the assumption of randomness", "abstract": "Conformal predictors provide set or functional predictions that are valid under the assumption of randomness, i.e., under the assumption of independent and identically distributed data. The question asked in this paper is whether there are predictors that are valid in the same sense under the assumption of randomness and that are more efficient than conformal predictors. The answer is that the class of conformal predictors is universal in that only limited gains in predictive efficiency are possible. The previous work in this area has relied on the algorithmic theory of randomness and so involved unspecified constants, whereas this paper's results are much more practical. They are also shown to be optimal in some respects.", "date": "2025-02-26", "categories": []} {"id": "2506.06599", "title": "Direct Prediction Set Minimization via Bilevel Conformal Classifier Training", "abstract": "Conformal prediction (CP) is a promising uncertainty quantification framework which works as a wrapper around a black-box classifier to construct prediction sets (i.e., subset of candidate classes) with provable guarantees. However, standard calibration methods for CP tend to produce large prediction sets which makes them less useful in practice. This paper considers the problem of integrating conformal principles into the training process of deep classifiers to directly minimize the size of prediction sets. We formulate conformal training as a bilevel optimization problem and propose the {\\em Direct Prediction Set Minimization (DPSM)} algorithm to solve it. The key insight behind DPSM is to minimize a measure of the prediction set size (upper level) that is conditioned on the learned quantile of conformity scores (lower level). We analyze that DPSM has a learning bound of $O(1/\\sqrt{n})$ (with $n$ training samples), while prior conformal training methods based on stochastic approximation for the quantile has a bound of $\\Omega(1/s)$ (with batch size $s$ and typically $s \\ll \\sqrt{n}$). Experiments on various benchmark datasets and deep models show that DPSM significantly outperforms the best prior conformal training baseline with $20.46\\%\\downarrow$ in the prediction set size and validates our theory.", "date": "2025-06-07", "categories": []} {"id": "2506.07804", "title": "Enhancing Adversarial Robustness with Conformal Prediction: A Framework for Guaranteed Model Reliability", "abstract": "As deep learning models are increasingly deployed in high-risk applications, robust defenses against adversarial attacks and reliable performance guarantees become paramount. Moreover, accuracy alone does not provide sufficient assurance or reliable uncertainty estimates for these models. This study advances adversarial training by leveraging principles from Conformal Prediction. Specifically, we develop an adversarial attack method, termed OPSA (OPtimal Size Attack), designed to reduce the efficiency of conformal prediction at any significance level by maximizing model uncertainty without requiring coverage guarantees. Correspondingly, we introduce OPSA-AT (Adversarial Training), a defense strategy that integrates OPSA within a novel conformal training paradigm. Experimental evaluations demonstrate that our OPSA attack method induces greater uncertainty compared to baseline approaches for various defenses. Conversely, our OPSA-AT defensive model significantly enhances robustness not only against OPSA but also other adversarial attacks, and maintains reliable prediction. Our findings highlight the effectiveness of this integrated approach for developing trustworthy and resilient deep learning models for safety-critical domains. Our code is available at https://github.com/bjbbbb/Enhancing-Adversarial-Robustness-with-Conformal-Prediction.", "date": "2025-06-09", "categories": []} {"id": "2506.07854", "title": "Residual Reweighted Conformal Prediction for Graph Neural Networks", "abstract": "Graph Neural Networks (GNNs) excel at modeling relational data but face significant challenges in high-stakes domains due to unquantified uncertainty. Conformal prediction (CP) offers statistical coverage guarantees, but existing methods often produce overly conservative prediction intervals that fail to account for graph heteroscedasticity and structural biases. While residual reweighting CP variants address some of these limitations, they neglect graph topology, cluster-specific uncertainties, and risk data leakage by reusing training sets. To address these issues, we propose Residual Reweighted GNN (RR-GNN), a framework designed to generate minimal prediction sets with provable marginal coverage guarantees.\n RR-GNN introduces three major innovations to enhance prediction performance. First, it employs Graph-Structured Mondrian CP to partition nodes or edges into communities based on topological features, ensuring cluster-conditional coverage that reflects heterogeneity. Second, it uses Residual-Adaptive Nonconformity Scores by training a secondary GNN on a held-out calibration set to estimate task-specific residuals, dynamically adjusting prediction intervals according to node or edge uncertainty. Third, it adopts a Cross-Training Protocol, which alternates the optimization of the primary GNN and the residual predictor to prevent information leakage while maintaining graph dependencies. We validate RR-GNN on 15 real-world graphs across diverse tasks, including node classification, regression, and edge weight prediction. Compared to CP baselines, RR-GNN achieves improved efficiency over state-of-the-art methods, with no loss of coverage.", "date": "2025-06-09", "categories": []} {"id": "2506.05497", "title": "Conformal Prediction Beyond the Seen: A Missing Mass Perspective for Uncertainty Quantification in Generative Models", "abstract": "Uncertainty quantification (UQ) is essential for safe deployment of generative AI models such as large language models (LLMs), especially in high stakes applications. Conformal prediction (CP) offers a principled uncertainty quantification framework, but classical methods focus on regression and classification, relying on geometric distances or softmax scores: tools that presuppose structured outputs. We depart from this paradigm by studying CP in a query only setting, where prediction sets must be constructed solely from finite queries to a black box generative model, introducing a new trade off between coverage, test time query budget, and informativeness. We introduce Conformal Prediction with Query Oracle (CPQ), a framework characterizing the optimal interplay between these objectives. Our finite sample algorithm is built on two core principles: one governs the optimal query policy, and the other defines the optimal mapping from queried samples to prediction sets. Remarkably, both are rooted in the classical missing mass problem in statistics. Specifically, the optimal query policy depends on the rate of decay, or the derivative, of the missing mass, for which we develop a novel estimator. Meanwhile, the optimal mapping hinges on the missing mass itself, which we estimate using Good Turing estimators. We then turn our focus to implementing our method for language models, where outputs are vast, variable, and often under specified. Fine grained experiments on three real world open ended tasks and two LLMs, show CPQ applicability to any black box LLM and highlight: (1) individual contribution of each principle to CPQ performance, and (2) CPQ ability to yield significantly more informative prediction sets than existing conformal methods for language uncertainty quantification.", "date": "2025-06-05", "categories": []} {"id": "2402.05806", "title": "On Temperature Scaling and Conformal Prediction of Deep Classifiers", "abstract": "In many classification applications, the prediction of a deep neural network (DNN) based classifier needs to be accompanied by some confidence indication. Two popular approaches for that aim are: 1) Calibration: modifies the classifier's softmax values such that the maximal value better estimates the correctness probability; and 2) Conformal Prediction (CP): produces a prediction set of candidate labels that contains the true label with a user-specified probability, guaranteeing marginal coverage but not, e.g., per class coverage. In practice, both types of indications are desirable, yet, so far the interplay between them has not been investigated. Focusing on the ubiquitous Temperature Scaling (TS) calibration, we start this paper with an extensive empirical study of its effect on prominent CP methods. We show that while TS calibration improves the class-conditional coverage of adaptive CP methods, surprisingly, it negatively affects their prediction set sizes. Motivated by this behavior, we explore the effect of TS on CP beyond its calibration application and reveal an intriguing trend under which it allows to trade prediction set size and conditional coverage of adaptive CP methods. Then, we establish a mathematical theory that explains the entire non-monotonic trend. Finally, based on our experiments and theory, we offer simple guidelines for practitioners to effectively combine adaptive CP with calibration, aligned with user-defined goals.", "date": "2024-02-08", "categories": []} {"id": "2505.24693", "title": "Conformal Prediction for Zero-Shot Models", "abstract": "Vision-language models pre-trained at large scale have shown unprecedented adaptability and generalization to downstream tasks. Although its discriminative potential has been widely explored, its reliability and uncertainty are still overlooked. In this work, we investigate the capabilities of CLIP models under the split conformal prediction paradigm, which provides theoretical guarantees to black-box models based on a small, labeled calibration set. In contrast to the main body of literature on conformal predictors in vision classifiers, foundation models exhibit a particular characteristic: they are pre-trained on a one-time basis on an inaccessible source domain, different from the transferred task. This domain drift negatively affects the efficiency of the conformal sets and poses additional challenges. To alleviate this issue, we propose Conf-OT, a transfer learning setting that operates transductive over the combined calibration and query sets. Solving an optimal transport problem, the proposed method bridges the domain gap between pre-training and adaptation without requiring additional data splits but still maintaining coverage guarantees. We comprehensively explore this conformal prediction strategy on a broad span of 15 datasets and three non-conformity scores. Conf-OT provides consistent relative improvements of up to 20% on set efficiency while being 15 times faster than popular transductive approaches.", "date": "2025-05-30", "categories": []} {"id": "2407.06390", "title": "JANET: Joint Adaptive predictioN-region Estimation for Time-series", "abstract": "Conformal prediction provides machine learning models with prediction sets that offer theoretical guarantees, but the underlying assumption of exchangeability limits its applicability to time series data. Furthermore, existing approaches struggle to handle multi-step ahead prediction tasks, where uncertainty estimates across multiple future time points are crucial. We propose JANET (Joint Adaptive predictioN-region Estimation for Time-series), a novel framework for constructing conformal prediction regions that are valid for both univariate and multivariate time series. JANET generalises the inductive conformal framework and efficiently produces joint prediction regions with controlled K-familywise error rates, enabling flexible adaptation to specific application needs. Our empirical evaluation demonstrates JANET's superior performance in multi-step prediction tasks across diverse time series datasets, highlighting its potential for reliable and interpretable uncertainty quantification in sequential data.", "date": "2024-07-08", "categories": []} {"id": "2505.22764", "title": "Test-time augmentation improves efficiency in conformal prediction", "abstract": "A conformal classifier produces a set of predicted classes and provides a probabilistic guarantee that the set includes the true class. Unfortunately, it is often the case that conformal classifiers produce uninformatively large sets. In this work, we show that test-time augmentation (TTA)--a technique that introduces inductive biases during inference--reduces the size of the sets produced by conformal classifiers. Our approach is flexible, computationally efficient, and effective. It can be combined with any conformal score, requires no model retraining, and reduces prediction set sizes by 10%-14% on average. We conduct an evaluation of the approach spanning three datasets, three models, two established conformal scoring methods, different guarantee strengths, and several distribution shifts to show when and why test-time augmentation is a useful addition to the conformal pipeline.", "date": "2025-05-28", "categories": []} {"id": "2505.23196", "title": "JAPAN: Joint Adaptive Prediction Areas with Normalising-Flows", "abstract": "Conformal prediction provides a model-agnostic framework for uncertainty quantification with finite-sample validity guarantees, making it an attractive tool for constructing reliable prediction sets. However, existing approaches commonly rely on residual-based conformity scores, which impose geometric constraints and struggle when the underlying distribution is multimodal. In particular, they tend to produce overly conservative prediction areas centred around the mean, often failing to capture the true shape of complex predictive distributions. In this work, we introduce JAPAN (Joint Adaptive Prediction Areas with Normalising-Flows), a conformal prediction framework that uses density-based conformity scores. By leveraging flow-based models, JAPAN estimates the (predictive) density and constructs prediction areas by thresholding on the estimated density scores, enabling compact, potentially disjoint, and context-adaptive regions that retain finite-sample coverage guarantees. We theoretically motivate the efficiency of JAPAN and empirically validate it across multivariate regression and forecasting tasks, demonstrating good calibration and tighter prediction areas compared to existing baselines. We also provide several \\emph{extensions} adding flexibility to our proposed framework.", "date": "2025-05-29", "categories": []} {"id": "2505.23389", "title": "Dynamic Estimation Loss Control in Variational Quantum Sensing via Online Conformal Inference", "abstract": "Quantum sensing exploits non-classical effects to overcome limitations of classical sensors, with applications ranging from gravitational-wave detection to nanoscale imaging. However, practical quantum sensors built on noisy intermediate-scale quantum (NISQ) devices face significant noise and sampling constraints, and current variational quantum sensing (VQS) methods lack rigorous performance guarantees. This paper proposes an online control framework for VQS that dynamically updates the variational parameters while providing deterministic error bars on the estimates. By leveraging online conformal inference techniques, the approach produces sequential estimation sets with a guaranteed long-term risk level. Experiments on a quantum magnetometry task confirm that the proposed dynamic VQS approach maintains the required reliability over time, while still yielding precise estimates. The results demonstrate the practical benefits of combining variational quantum algorithms with online conformal inference to achieve reliable quantum sensing on NISQ devices.", "date": "2025-05-29", "categories": []} {"id": "2405.13268", "title": "Stochastic Online Conformal Prediction with Semi-Bandit Feedback", "abstract": "Conformal prediction has emerged as an effective strategy for uncertainty quantification by modifying a model to output sets of labels instead of a single label. These prediction sets come with the guarantee that they contain the true label with high probability. However, conformal prediction typically requires a large calibration dataset of i.i.d. examples. We consider the online learning setting, where examples arrive over time, and the goal is to construct prediction sets dynamically. Departing from existing work, we assume semi-bandit feedback, where we only observe the true label if it is contained in the prediction set. For instance, consider calibrating a document retrieval model to a new domain; in this setting, a user would only be able to provide the true label if the target document is in the prediction set of retrieved documents. We propose a novel conformal prediction algorithm targeted at this setting, and prove that it obtains sublinear regret compared to the optimal conformal predictor. We evaluate our algorithm on a retrieval task, an image classification task, and an auction price-setting task, and demonstrate that it empirically achieves good performance compared to several baselines.", "date": "2024-05-22", "categories": []} {"id": "2505.21039", "title": "Scalable and adaptive prediction bands with kernel sum-of-squares", "abstract": "Conformal Prediction (CP) is a popular framework for constructing prediction bands with valid coverage in finite samples, while being free of any distributional assumption. A well-known limitation of conformal prediction is the lack of adaptivity, although several works introduced practically efficient alternate procedures. In this work, we build upon recent ideas that rely on recasting the CP problem as a statistical learning problem, directly targeting coverage and adaptivity. This statistical learning problem is based on reproducible kernel Hilbert spaces (RKHS) and kernel sum-of-squares (SoS) methods. First, we extend previous results with a general representer theorem and exhibit the dual formulation of the learning problem. Crucially, such dual formulation can be solved efficiently by accelerated gradient methods with several hundreds or thousands of samples, unlike previous strategies based on off-the-shelf semidefinite programming algorithms. Second, we introduce a new hyperparameter tuning strategy tailored specifically to target adaptivity through bounds on test-conditional coverage. This strategy, based on the Hilbert-Schmidt Independence Criterion (HSIC), is introduced here to tune kernel lengthscales in our framework, but has broader applicability since it could be used in any CP algorithm where the score function is learned. Finally, extensive experiments are conducted to show how our method compares to related work. All figures can be reproduced with the accompanying code.", "date": "2025-05-27", "categories": []} {"id": "2505.18890", "title": "Conformal Prediction for Uncertainty Estimation in Drug-Target Interaction Prediction", "abstract": "Accurate drug-target interaction (DTI) prediction with machine learning models is essential for drug discovery. Such models should also provide a credible representation of their uncertainty, but applying classical marginal conformal prediction (CP) in DTI prediction often overlooks variability across drug and protein subgroups. In this work, we analyze three cluster-conditioned CP methods for DTI prediction, and compare them with marginal and group-conditioned CP. Clusterings are obtained via nonconformity scores, feature similarity, and nearest neighbors, respectively. Experiments on the KIBA dataset using four data-splitting strategies show that nonconformity-based clustering yields the tightest intervals and most reliable subgroup coverage, especially in random and fully unseen drug-protein splits. Group-conditioned CP works well when one entity is familiar, but residual-driven clustering provides robust uncertainty estimates even in sparse or novel scenarios. These results highlight the potential of cluster-based CP for improving DTI prediction under uncertainty.", "date": "2025-05-24", "categories": []} {"id": "2505.19033", "title": "Optimal Conformal Prediction under Epistemic Uncertainty", "abstract": "Conformal prediction (CP) is a popular frequentist framework for representing uncertainty by providing prediction sets that guarantee coverage of the true label with a user-adjustable probability. In most applications, CP operates on confidence scores coming from a standard (first-order) probabilistic predictor (e.g., softmax outputs). Second-order predictors, such as credal set predictors or Bayesian models, are also widely used for uncertainty quantification and are known for their ability to represent both aleatoric and epistemic uncertainty. Despite their popularity, there is still an open question on ``how they can be incorporated into CP''. In this paper, we discuss the desiderata for CP when valid second-order predictions are available. We then introduce Bernoulli prediction sets (BPS), which produce the smallest prediction sets that ensure conditional coverage in this setting. When given first-order predictions, BPS reduces to the well-known adaptive prediction sets (APS). Furthermore, when the validity assumption on the second-order predictions is compromised, we apply conformal risk control to obtain a marginal coverage guarantee while still accounting for epistemic uncertainty.", "date": "2025-05-25", "categories": []} {"id": "2505.19587", "title": "WQLCP: Weighted Adaptive Conformal Prediction for Robust Uncertainty Quantification Under Distribution Shifts", "abstract": "Conformal prediction (CP) provides a framework for constructing prediction sets with guaranteed coverage, assuming exchangeable data. However, real-world scenarios often involve distribution shifts that violate exchangeability, leading to unreliable coverage and inflated prediction sets. To address this challenge, we first introduce Reconstruction Loss-Scaled Conformal Prediction (RLSCP), which utilizes reconstruction losses derived from a Variational Autoencoder (VAE) as an uncertainty metric to scale score functions. While RLSCP demonstrates performance improvements, mainly resulting in better coverage, it quantifies quantiles based on a fixed calibration dataset without considering the discrepancies between test and train datasets in an unexchangeable setting. In the next step, we propose Weighted Quantile Loss-scaled Conformal Prediction (WQLCP), which refines RLSCP by incorporating a weighted notion of exchangeability, adjusting the calibration quantile threshold based on weights with respect to the ratio of calibration and test loss values. This approach improves the CP-generated prediction set outputs in the presence of distribution shifts. Experiments on large-scale datasets, including ImageNet variants, demonstrate that WQLCP outperforms existing baselines by consistently maintaining coverage while reducing prediction set sizes, providing a robust solution for CP under distribution shifts.", "date": "2025-05-26", "categories": []} {"id": "2412.09729", "title": "Doubly Robust Conformalized Survival Analysis with Right-Censored Data", "abstract": "We present a conformal inference method for constructing lower prediction bounds for survival times from right-censored data, extending recent approaches designed for more restrictive type-I censoring scenarios. The proposed method imputes unobserved censoring times using a machine learning model, and then analyzes the imputed data using a survival model calibrated via weighted conformal inference. This approach is theoretically supported by an asymptotic double robustness property. Empirical studies on simulated and real data demonstrate that our method leads to relatively informative predictive inferences and is especially robust in challenging settings where the survival model may be inaccurate.", "date": "2024-12-12", "categories": []} {"id": "2505.17126", "title": "Conformal Language Model Reasoning with Coherent Factuality", "abstract": "Language models are increasingly being used in important decision pipelines, so ensuring the correctness of their outputs is crucial. Recent work has proposed evaluating the \"factuality\" of claims decomposed from a language model generation and applying conformal prediction techniques to filter out those claims that are not factual. This can be effective for tasks such as information retrieval, where constituent claims may be evaluated in isolation for factuality, but is not appropriate for reasoning tasks, as steps of a logical argument can be evaluated for correctness only within the context of the claims that precede them. To capture this, we define \"coherent factuality\" and develop a conformal-prediction-based method to guarantee coherent factuality for language model outputs. Our approach applies split conformal prediction to subgraphs within a \"deducibility\" graph\" that represents the steps of a reasoning problem. We evaluate our method on mathematical reasoning problems from the MATH and FELM datasets and find that our algorithm consistently produces correct and substantiated orderings of claims, achieving coherent factuality across target coverage levels. Moreover, we achieve 90% factuality on our stricter definition while retaining 80% or more of the original claims, highlighting the utility of our deducibility-graph-guided approach.", "date": "2025-05-21", "categories": []} {"id": "2410.02561", "title": "The Benefit of Being Bayesian in Online Conformal Prediction", "abstract": "Based on the framework of Conformal Prediction (CP), we study the online construction of confidence sets given a black-box machine learning model. By converting the target confidence levels into quantile levels, the problem can be reduced to predicting the quantiles (in hindsight) of a sequentially revealed data sequence. Two very different approaches have been studied previously: (i) Assuming the data sequence is iid or exchangeable, one could maintain the empirical distribution of the observed data as an algorithmic belief, and directly predict its quantiles. (ii) Due to the fragility of statistical assumptions, a recent trend is to consider the non-distributional, adversarial setting and apply first-order online optimization algorithms to moving quantile losses. However, it requires the oracle knowledge of the target quantile level, and suffers from a previously overlooked monotonicity issue due to the associated loss linearization.\n This paper presents an adaptive CP algorithm that combines their strengths. Without any statistical assumption, it is able to answer multiple arbitrary confidence level queries with low regret, while also overcoming the monotonicity issue suffered by first-order optimization baselines. Furthermore, if the data sequence is actually iid, then the same algorithm is automatically equipped with the \"correct\" coverage probability guarantee.\n To achieve such strengths, our key technical innovation is to regularize the aforementioned algorithmic belief (the empirical distribution) by a Bayesian prior, which robustifies it by simulating a non-linearized Follow the Regularized Leader (FTRL) algorithm on the output. Such a belief update backbone is shared by prediction heads targeting different confidence levels, bringing practical benefits analogous to the recently proposed concept of U-calibration (Kleinberg et al., 2023).", "date": "2024-10-03", "categories": []} {"id": "2501.14544", "title": "Distributed Conformal Prediction via Message Passing", "abstract": "Post-hoc calibration of pre-trained models is critical for ensuring reliable inference, especially in safety-critical domains such as healthcare. Conformal Prediction (CP) offers a robust post-hoc calibration framework, providing distribution-free statistical coverage guarantees for prediction sets by leveraging held-out datasets. In this work, we address a decentralized setting where each device has limited calibration data and can communicate only with its neighbors over an arbitrary graph topology. We propose two message-passing-based approaches for achieving reliable inference via CP: quantile-based distributed conformal prediction (Q-DCP) and histogram-based distributed conformal prediction (H-DCP). Q-DCP employs distributed quantile regression enhanced with tailored smoothing and regularization terms to accelerate convergence, while H-DCP uses a consensus-based histogram estimation approach. Through extensive experiments, we investigate the trade-offs between hyperparameter tuning requirements, communication overhead, coverage guarantees, and prediction set sizes across different network topologies. The code of our work is released on: https://github.com/HaifengWen/Distributed-Conformal-Prediction.", "date": "2025-01-24", "categories": []} {"id": "2503.01495", "title": "Improving the statistical efficiency of cross-conformal prediction", "abstract": "Vovk (2015) introduced cross-conformal prediction, a modification of split conformal designed to improve the width of prediction sets. The method, when trained with a miscoverage rate equal to $\\alpha$ and $n \\gg K$, ensures a marginal coverage of at least $1 - 2\\alpha - 2(1-\\alpha)(K-1)/(n+K)$, where $n$ is the number of observations and $K$ denotes the number of folds. A simple modification of the method achieves coverage of at least $1-2\\alpha$. In this work, we propose new variants of both methods that yield smaller prediction sets without compromising the latter theoretical guarantees. The proposed methods are based on recent results deriving more statistically efficient combination of p-values that leverage exchangeability and randomization. Simulations confirm the theoretical findings and bring out some important tradeoffs.", "date": "2025-03-03", "categories": []} {"id": "2409.18332", "title": "Conformal Prediction: A Theoretical Note and Benchmarking Transductive Node Classification in Graphs", "abstract": "Conformal prediction has become increasingly popular for quantifying the uncertainty associated with machine learning models. Recent work in graph uncertainty quantification has built upon this approach for conformal graph prediction. The nascent nature of these explorations has led to conflicting choices for implementations, baselines, and method evaluation. In this work, we analyze the design choices made in the literature and discuss the tradeoffs associated with existing methods. Building on the existing implementations, we introduce techniques to scale existing methods to large-scale graph datasets without sacrificing performance. Our theoretical and empirical results justify our recommendations for future scholarship in graph conformal prediction.", "date": "2024-09-26", "categories": []} {"id": "2502.14105", "title": "Conformal Prediction under Levy-Prokhorov Distribution Shifts: Robustness to Local and Global Perturbations", "abstract": "Conformal prediction provides a powerful framework for constructing prediction intervals with finite-sample guarantees, yet its robustness under distribution shifts remains a significant challenge. This paper addresses this limitation by modeling distribution shifts using Levy-Prokhorov (LP) ambiguity sets, which capture both local and global perturbations. We provide a self-contained overview of LP ambiguity sets and their connections to popular metrics such as Wasserstein and Total Variation. We show that the link between conformal prediction and LP ambiguity sets is a natural one: by propagating the LP ambiguity set through the scoring function, we reduce complex high-dimensional distribution shifts to manageable one-dimensional distribution shifts, enabling exact quantification of worst-case quantiles and coverage. Building on this analysis, we construct robust conformal prediction intervals that remain valid under distribution shifts, explicitly linking LP parameters to interval width and confidence levels. Experimental results on real-world datasets demonstrate the effectiveness of the proposed approach.", "date": "2025-02-19", "categories": []} {"id": "2504.17671", "title": "Data-Driven Calibration of Prediction Sets in Large Vision-Language Models Based on Inductive Conformal Prediction", "abstract": "This study addresses the critical challenge of hallucination mitigation in Large Vision-Language Models (LVLMs) for Visual Question Answering (VQA) tasks through a Split Conformal Prediction (SCP) framework. While LVLMs excel in multi-modal reasoning, their outputs often exhibit hallucinated content with high confidence, posing risks in safety-critical applications. We propose a model-agnostic uncertainty quantification method that integrates dynamic threshold calibration and cross-modal consistency verification. By partitioning data into calibration and test sets, the framework computes nonconformity scores to construct prediction sets with statistical guarantees under user-defined risk levels ($\\alpha$). Key innovations include: (1) rigorous control of \\textbf{marginal coverage} to ensure empirical error rates remain strictly below $\\alpha$; (2) dynamic adjustment of prediction set sizes inversely with $\\alpha$, filtering low-confidence outputs; (3) elimination of prior distribution assumptions and retraining requirements. Evaluations on benchmarks (ScienceQA, MMMU) with eight LVLMs demonstrate that SCP enforces theoretical guarantees across all $\\alpha$ values. The framework achieves stable performance across varying calibration-to-test split ratios, underscoring its robustness for real-world deployment in healthcare, autonomous systems, and other safety-sensitive domains. This work bridges the gap between theoretical reliability and practical applicability in multi-modal AI systems, offering a scalable solution for hallucination detection and uncertainty-aware decision-making.", "date": "2025-04-24", "categories": []} {"id": "2505.08845", "title": "Validation of Conformal Prediction in Cervical Atypia Classification", "abstract": "Deep learning based cervical cancer classification can potentially increase access to screening in low-resource regions. However, deep learning models are often overconfident and do not reliably reflect diagnostic uncertainty. Moreover, they are typically optimized to generate maximum-likelihood predictions, which fail to convey uncertainty or ambiguity in their results. Such challenges can be addressed using conformal prediction, a model-agnostic framework for generating prediction sets that contain likely classes for trained deep-learning models. The size of these prediction sets indicates model uncertainty, contracting as model confidence increases. However, existing conformal prediction evaluation primarily focuses on whether the prediction set includes or covers the true class, often overlooking the presence of extraneous classes. We argue that prediction sets should be truthful and valuable to end users, ensuring that the listed likely classes align with human expectations rather than being overly relaxed and including false positives or unlikely classes. In this study, we comprehensively validate conformal prediction sets using expert annotation sets collected from multiple annotators. We evaluate three conformal prediction approaches applied to three deep-learning models trained for cervical atypia classification. Our expert annotation-based analysis reveals that conventional coverage-based evaluations overestimate performance and that current conformal prediction methods often produce prediction sets that are not well aligned with human labels. Additionally, we explore the capabilities of the conformal prediction methods in identifying ambiguous and out-of-distribution data.", "date": "2025-05-13", "categories": []} {"id": "2505.08158", "title": "Feature Fitted Online Conformal Prediction for Deep Time Series Forecasting Model", "abstract": "Time series forecasting is critical for many applications, where deep learning-based point prediction models have demonstrated strong performance. However, in practical scenarios, there is also a need to quantify predictive uncertainty through online confidence intervals. Existing confidence interval modeling approaches building upon these deep point prediction models suffer from key limitations: they either require costly retraining, fail to fully leverage the representational strengths of deep models, or lack theoretical guarantees. To address these gaps, we propose a lightweight conformal prediction method that provides valid coverage and shorter interval lengths without retraining. Our approach leverages features extracted from pre-trained point prediction models to fit a residual predictor and construct confidence intervals, further enhanced by an adaptive coverage control mechanism. Theoretically, we prove that our method achieves asymptotic coverage convergence, with error bounds dependent on the feature quality of the underlying point prediction model. Experiments on 12 datasets demonstrate that our method delivers tighter confidence intervals while maintaining desired coverage rates. Code, model and dataset in \\href{https://github.com/xiannanhuang/FFDCI}{Github}", "date": "2025-05-13", "categories": []} {"id": "2503.05505", "title": "Correctness Coverage Evaluation for Medical Multiple-Choice Question\n Answering Based on the Enhanced Conformal Prediction Framework", "abstract": " Large language models (LLMs) are increasingly adopted in medical\nquestion-answering (QA) scenarios. However, LLMs can generate hallucinations\nand nonfactual information, undermining their trustworthiness in high-stakes\nmedical tasks. Conformal Prediction (CP) provides a statistically rigorous\nframework for marginal (average) coverage guarantees but has limited\nexploration in medical QA. This paper proposes an enhanced CP framework for\nmedical multiple-choice question-answering (MCQA) tasks. By associating the\nnon-conformance score with the frequency score of correct options and\nleveraging self-consistency, the framework addresses internal model opacity and\nincorporates a risk control strategy with a monotonic loss function. Evaluated\non MedMCQA, MedQA, and MMLU datasets using four off-the-shelf LLMs, the\nproposed method meets specified error rate guarantees while reducing average\nprediction set size with increased risk level, offering a promising uncertainty\nevaluation metric for LLMs.\n", "date": "2025-03-07", "categories": []} {"id": "2505.04986", "title": "Conformal Prediction with Cellwise Outliers: A Detect-then-Impute\n Approach", "abstract": " Conformal prediction is a powerful tool for constructing prediction intervals\nfor black-box models, providing a finite sample coverage guarantee for\nexchangeable data. However, this exchangeability is compromised when some\nentries of the test feature are contaminated, such as in the case of cellwise\noutliers. To address this issue, this paper introduces a novel framework called\ndetect-then-impute conformal prediction. This framework first employs an\noutlier detection procedure on the test feature and then utilizes an imputation\nmethod to fill in those cells identified as outliers. To quantify the\nuncertainty in the processed test feature, we adaptively apply the detection\nand imputation procedures to the calibration set, thereby constructing\nexchangeable features for the conformal prediction interval of the test label.\nWe develop two practical algorithms, PDI-CP and JDI-CP, and provide a\ndistribution-free coverage analysis under some commonly used detection and\nimputation procedures. Notably, JDI-CP achieves a finite sample $1-2\\alpha$\ncoverage guarantee. Numerical experiments on both synthetic and real datasets\ndemonstrate that our proposed algorithms exhibit robust coverage properties and\ncomparable efficiency to the oracle baseline.\n", "date": "2025-05-08", "categories": []} {"id": "2503.22712", "title": "Coverage-Guaranteed Speech Emotion Recognition via Calibrated\n Uncertainty-Adaptive Prediction Sets", "abstract": " Road rage, often triggered by emotional suppression and sudden outbursts,\nsignificantly threatens road safety by causing collisions and aggressive\nbehavior. Speech emotion recognition technologies can mitigate this risk by\nidentifying negative emotions early and issuing timely alerts. However, current\nSER methods, such as those based on hidden markov models and Long short-term\nmemory networks, primarily handle one-dimensional signals, frequently\nexperience overfitting, and lack calibration, limiting their safety-critical\neffectiveness. We propose a novel risk-controlled prediction framework\nproviding statistically rigorous guarantees on prediction accuracy. This\napproach employs a calibration set to define a binary loss function indicating\nwhether the true label is included in the prediction set. Using a data-driven\nthreshold $\\beta$, we optimize a joint loss function to maintain an expected\ntest loss bounded by a user-specified risk level $\\alpha$. Evaluations across\nsix baseline models and two benchmark datasets demonstrate our framework\nconsistently achieves a minimum coverage of $1 - \\alpha$, effectively\ncontrolling marginal error rates despite varying calibration-test split ratios\n(e.g., 0.1). The robustness and generalizability of the framework are further\nvalidated through an extension to small-batch online calibration under a local\nexchangeability assumption. We construct a non-negative test martingale to\nmaintain prediction validity even in dynamic and non-exchangeable environments.\nCross-dataset tests confirm our method's ability to uphold reliable statistical\nguarantees in realistic, evolving data scenarios.\n", "date": "2025-03-24", "categories": []} {"id": "2501.18991", "title": "Optimal Transport-based Conformal Prediction", "abstract": " Conformal Prediction (CP) is a principled framework for quantifying\nuncertainty in blackbox learning models, by constructing prediction sets with\nfinite-sample coverage guarantees. Traditional approaches rely on scalar\nnonconformity scores, which fail to fully exploit the geometric structure of\nmultivariate outputs, such as in multi-output regression or multiclass\nclassification. Recent methods addressing this limitation impose predefined\nconvex shapes for the prediction sets, potentially misaligning with the\nintrinsic data geometry. We introduce a novel CP procedure handling\nmultivariate score functions through the lens of optimal transport.\nSpecifically, we leverage Monge-Kantorovich vector ranks and quantiles to\nconstruct prediction region with flexible, potentially non-convex shapes,\nbetter suited to the complex uncertainty patterns encountered in multivariate\nlearning tasks. We prove that our approach ensures finite-sample,\ndistribution-free coverage properties, similar to typical CP methods. We then\nadapt our method for multi-output regression and multiclass classification, and\nalso propose simple adjustments to generate adaptive prediction regions with\nasymptotic conditional coverage guarantees. Finally, we evaluate our method on\npractical regression and classification problems, illustrating its advantages\nin terms of (conditional) coverage and efficiency.\n", "date": "2025-01-31", "categories": []} {"id": "2503.13050", "title": "E-Values Expand the Scope of Conformal Prediction", "abstract": " Conformal prediction is a powerful framework for distribution-free\nuncertainty quantification. The standard approach to conformal prediction\nrelies on comparing the ranks of prediction scores: under exchangeability, the\nrank of a future test point cannot be too extreme relative to a calibration\nset. This rank-based method can be reformulated in terms of p-values. In this\npaper, we explore an alternative approach based on e-values, known as conformal\ne-prediction. E-values offer key advantages that cannot be achieved with\np-values, enabling new theoretical and practical capabilities. In particular,\nwe present three applications that leverage the unique strengths of e-values:\nbatch anytime-valid conformal prediction, fixed-size conformal sets with\ndata-dependent coverage, and conformal prediction under ambiguous ground truth.\nOverall, these examples demonstrate that e-value-based constructions provide a\nflexible expansion of the toolbox of conformal prediction.\n", "date": "2025-03-17", "categories": []} {"id": "2505.01783", "title": "Context-Aware Online Conformal Anomaly Detection with Prediction-Powered\n Data Acquisition", "abstract": " Online anomaly detection is essential in fields such as cybersecurity,\nhealthcare, and industrial monitoring, where promptly identifying deviations\nfrom expected behavior can avert critical failures or security breaches. While\nnumerous anomaly scoring methods based on supervised or unsupervised learning\nhave been proposed, current approaches typically rely on a continuous stream of\nreal-world calibration data to provide assumption-free guarantees on the false\ndiscovery rate (FDR). To address the inherent challenges posed by limited real\ncalibration data, we introduce context-aware prediction-powered conformal\nonline anomaly detection (C-PP-COAD). Our framework strategically leverages\nsynthetic calibration data to mitigate data scarcity, while adaptively\nintegrating real data based on contextual cues. C-PP-COAD utilizes conformal\np-values, active p-value statistics, and online FDR control mechanisms to\nmaintain rigorous and reliable anomaly detection performance over time.\nExperiments conducted on both synthetic and real-world datasets demonstrate\nthat C-PP-COAD significantly reduces dependency on real calibration data\nwithout compromising guaranteed FDR control.\n", "date": "2025-05-03", "categories": []} {"id": "2312.08999", "title": "Conformalised data synthesis", "abstract": " With the proliferation of increasingly complicated Deep Learning\narchitectures, data synthesis is a highly promising technique to address the\ndemand of data-hungry models. However, reliably assessing the quality of a\n'synthesiser' model's output is an open research question with significant\nassociated risks for high-stake domains. To address this challenge, we propose\na unique synthesis algorithm that generates data from high-confidence feature\nspace regions based on the Conformal Prediction framework. We support our\nproposed algorithm with a comprehensive exploration of the core parameter's\ninfluence, an in-depth discussion of practical advice, and an extensive\nempirical evaluation of five benchmark datasets. To show our approach's\nversatility on ubiquitous real-world challenges, the datasets were carefully\nselected for their variety of difficult characteristics: low sample count,\nclass imbalance, and non-separability. In all trials, training sets extended\nwith our confident synthesised data performed at least as well as the original\nset and frequently significantly improved Deep Learning performance by up to 61\npercentage points F1-score.\n", "date": "2023-12-14", "categories": []} {"id": "2505.00917", "title": "Multivariate Conformal Selection", "abstract": " Selecting high-quality candidates from large datasets is critical in\napplications such as drug discovery, precision medicine, and alignment of large\nlanguage models (LLMs). While Conformal Selection (CS) provides rigorous\nuncertainty quantification, it is limited to univariate responses and scalar\ncriteria. To address this issue, we propose Multivariate Conformal Selection\n(mCS), a generalization of CS designed for multivariate response settings. Our\nmethod introduces regional monotonicity and employs multivariate nonconformity\nscores to construct conformal p-values, enabling finite-sample False Discovery\nRate (FDR) control. We present two variants: mCS-dist, using distance-based\nscores, and mCS-learn, which learns optimal scores via differentiable\noptimization. Experiments on simulated and real-world datasets demonstrate that\nmCS significantly improves selection power while maintaining FDR control,\nestablishing it as a robust framework for multivariate selection tasks.\n", "date": "2025-05-01", "categories": []} {"id": "2412.08661", "title": "GeoConformal prediction: a model-agnostic framework of measuring the\n uncertainty of spatial prediction", "abstract": " Spatial prediction is a fundamental task in geography. In recent years, with\nadvances in geospatial artificial intelligence (GeoAI), numerous models have\nbeen developed to improve the accuracy of geographic variable predictions.\nBeyond achieving higher accuracy, it is equally important to obtain predictions\nwith uncertainty measures to enhance model credibility and support responsible\nspatial prediction. Although geostatistic methods like Kriging offer some level\nof uncertainty assessment, such as Kriging variance, these measurements are not\nalways accurate and lack general applicability to other spatial models. To\naddress this issue, we propose a model-agnostic uncertainty assessment method\ncalled GeoConformal Prediction, which incorporates geographical weighting into\nconformal prediction. We applied it to two classic spatial prediction cases,\nspatial regression and spatial interpolation, to evaluate its reliability.\nFirst, in the spatial regression case, we used XGBoost to predict housing\nprices, followed by GeoConformal to calculate uncertainty. Our results show\nthat GeoConformal achieved a coverage rate of 93.67%, while Bootstrap methods\nonly reached a maximum coverage of 81.00% after 2000 runs. Next, we applied\nGeoConformal to spatial interpolation models. We found that the uncertainty\nobtained from GeoConformal aligned closely with the variance in Kriging.\nFinally, using GeoConformal, we analyzed the sources of uncertainty in spatial\nprediction. We found that explicitly including local features in AI models can\nsignificantly reduce prediction uncertainty, especially in areas with strong\nlocal dependence. Our findings suggest that GeoConformal holds potential not\nonly for geographic knowledge discovery but also for guiding the design of\nfuture GeoAI models, paving the way for more reliable and interpretable spatial\nprediction frameworks.\n", "date": "2024-12-05", "categories": []} {"id": "2408.08990", "title": "Adaptive Uncertainty Quantification for Generative AI", "abstract": " This work is concerned with conformal prediction in contemporary applications\n(including generative AI) where a black-box model has been trained on data that\nare not accessible to the user. Mirroring split-conformal inference, we design\na wrapper around a black-box algorithm which calibrates conformity scores. This\ncalibration is local and proceeds in two stages by first adaptively\npartitioning the predictor space into groups and then calibrating sectionally\ngroup by group. Adaptive partitioning (self-grouping) is achieved by fitting a\nrobust regression tree to the conformity scores on the calibration set. This\nnew tree variant is designed in such a way that adding a single new observation\ndoes not change the tree fit with overwhelmingly large probability. This\nadd-one-in robustness property allows us to conclude a finite sample\ngroup-conditional coverage guarantee, a refinement of the marginal guarantee.\nIn addition, unlike traditional split-conformal inference, adaptive splitting\nand within-group calibration yields adaptive bands which can stretch and shrink\nlocally. We demonstrate benefits of local tightening on several simulated as\nwell as real examples using non-parametric regression. Finally, we consider two\ncontemporary classification applications for obtaining uncertainty\nquantification around GPT-4o predictions. We conformalize skin disease\ndiagnoses based on self-reported symptoms as well as predicted states of U.S.\nlegislators based on summaries of their ideology. We demonstrate substantial\nlocal tightening of the uncertainty sets while attaining similar marginal\ncoverage.\n", "date": "2024-08-16", "categories": []} {"id": "2501.06878", "title": "Uncertainty-Aware Online Extrinsic Calibration: A Conformal Prediction\n Approach", "abstract": " Accurate sensor calibration is crucial for autonomous systems, yet its\nuncertainty quantification remains underexplored. We present the first approach\nto integrate uncertainty awareness into online extrinsic calibration, combining\nMonte Carlo Dropout with Conformal Prediction to generate prediction intervals\nwith a guaranteed level of coverage. Our method proposes a framework to enhance\nexisting calibration models with uncertainty quantification, compatible with\nvarious network architectures. Validated on KITTI (RGB Camera-LiDAR) and DSEC\n(Event Camera-LiDAR) datasets, we demonstrate effectiveness across different\nvisual sensor types, measuring performance with adapted metrics to evaluate the\nefficiency and reliability of the intervals. By providing calibration\nparameters with quantifiable confidence measures, we offer insights into the\nreliability of calibration estimates, which can greatly improve the robustness\nof sensor fusion in dynamic environments and usefully serve the Computer Vision\ncommunity.\n", "date": "2025-01-12", "categories": []} {"id": "2504.17655", "title": "Aerial Image Classification in Scarce and Unconstrained Environments via\n Conformal Prediction", "abstract": " This paper presents a comprehensive empirical analysis of conformal\nprediction methods on a challenging aerial image dataset featuring diverse\nevents in unconstrained environments. Conformal prediction is a powerful\npost-hoc technique that takes the output of any classifier and transforms it\ninto a set of likely labels, providing a statistical guarantee on the coverage\nof the true label. Unlike evaluations on standard benchmarks, our study\naddresses the complexities of data-scarce and highly variable real-world\nsettings. We investigate the effectiveness of leveraging pretrained models\n(MobileNet, DenseNet, and ResNet), fine-tuned with limited labeled data, to\ngenerate informative prediction sets. To further evaluate the impact of\ncalibration, we consider two parallel pipelines (with and without temperature\nscaling) and assess performance using two key metrics: empirical coverage and\naverage prediction set size. This setup allows us to systematically examine how\ncalibration choices influence the trade-off between reliability and efficiency.\nOur findings demonstrate that even with relatively small labeled samples and\nsimple nonconformity scores, conformal prediction can yield valuable\nuncertainty estimates for complex tasks. Moreover, our analysis reveals that\nwhile temperature scaling is often employed for calibration, it does not\nconsistently lead to smaller prediction sets, underscoring the importance of\ncareful consideration in its application. Furthermore, our results highlight\nthe significant potential of model compression techniques within the conformal\nprediction pipeline for deployment in resource-constrained environments. Based\non our observations, we advocate for future research to delve into the impact\nof noisy or ambiguous labels on conformal prediction performance and to explore\neffective model reduction strategies.\n", "date": "2025-04-24", "categories": []} {"id": "2504.15722", "title": "From predictions to confidence intervals: an empirical study of\n conformal prediction methods for in-context learning", "abstract": " Transformers have become a standard architecture in machine learning,\ndemonstrating strong in-context learning (ICL) abilities that allow them to\nlearn from the prompt at inference time. However, uncertainty quantification\nfor ICL remains an open challenge, particularly in noisy regression tasks. This\npaper investigates whether ICL can be leveraged for distribution-free\nuncertainty estimation, proposing a method based on conformal prediction to\nconstruct prediction intervals with guaranteed coverage. While traditional\nconformal methods are computationally expensive due to repeated model fitting,\nwe exploit ICL to efficiently generate confidence intervals in a single forward\npass. Our empirical analysis compares this approach against ridge\nregression-based conformal methods, showing that conformal prediction with\nin-context learning (CP with ICL) achieves robust and scalable uncertainty\nestimates. Additionally, we evaluate its performance under distribution shifts\nand establish scaling laws to guide model training. These findings bridge ICL\nand conformal prediction, providing a theoretically grounded and new framework\nfor uncertainty quantification in transformer-based models.\n", "date": "2025-04-22", "categories": []} {"id": "2502.06331", "title": "Conformal Prediction Regions are Imprecise Highest Density Regions", "abstract": " Recently, Cella and Martin proved how, under an assumption called consonance,\na credal set (i.e. a closed and convex set of probabilities) can be derived\nfrom the conformal transducer associated with transductive conformal\nprediction. We show that the Imprecise Highest Density Region (IHDR) associated\nwith such a credal set corresponds to the classical Conformal Prediction\nRegion. In proving this result, we establish a new relationship between\nConformal Prediction and Imprecise Probability (IP) theories, via the IP\nconcept of a cloud. A byproduct of our presentation is the discovery that\nconsonant plausibility functions are monoid homomorphisms, a new algebraic\nproperty of an IP tool.\n", "date": "2025-02-10", "categories": []} {"id": "2504.12189", "title": "Leave-One-Out Stable Conformal Prediction", "abstract": " Conformal prediction (CP) is an important tool for distribution-free\npredictive uncertainty quantification. Yet, a major challenge is to balance\ncomputational efficiency and prediction accuracy, particularly for multiple\npredictions. We propose Leave-One-Out Stable Conformal Prediction (LOO-StabCP),\na novel method to speed up full conformal using algorithmic stability without\nsample splitting. By leveraging leave-one-out stability, our method is much\nfaster in handling a large number of prediction requests compared to existing\nmethod RO-StabCP based on replace-one stability. We derived stability bounds\nfor several popular machine learning tools: regularized loss minimization (RLM)\nand stochastic gradient descent (SGD), as well as kernel method, neural\nnetworks and bagging. Our method is theoretically justified and demonstrates\nsuperior numerical performance on synthetic and real-world data. We applied our\nmethod to a screening problem, where its effective exploitation of training\ndata led to improved test power compared to state-of-the-art method based on\nsplit conformal.\n", "date": "2025-04-16", "categories": []} {"id": "2503.05439", "title": "An Empirical Study of Conformal Prediction in LLM with ASP Scaffolds for\n Robust Reasoning", "abstract": " In this paper, we examine the use of Conformal Language Modelling (CLM)\nalongside Answer Set Programming (ASP) to enhance the performance of standard\nopen-weight LLMs on complex multi-step reasoning tasks. Using the StepGame\ndataset, which requires spatial reasoning, we apply CLM to generate sets of ASP\nprograms from an LLM, providing statistical guarantees on the correctness of\nthe outputs. Experimental results show that CLM significantly outperforms\nbaseline models that use standard sampling methods, achieving substantial\naccuracy improvements across different levels of reasoning complexity.\nAdditionally, the LLM-as-Judge metric enhances CLM's performance, especially in\nassessing structurally and logically correct ASP outputs. However, calibrating\nCLM with diverse calibration sets did not improve generalizability for tasks\nrequiring much longer reasoning steps, indicating limitations in handling more\ncomplex tasks.\n", "date": "2025-03-07", "categories": []} {"id": "2504.07611", "title": "Conditional Conformal Risk Adaptation", "abstract": " Uncertainty quantification is becoming increasingly important in image\nsegmentation, especially for high-stakes applications like medical imaging.\nWhile conformal risk control generalizes conformal prediction beyond standard\nmiscoverage to handle various loss functions such as false negative rate, its\napplication to segmentation often yields inadequate conditional risk control:\nsome images experience very high false negative rates while others have\nnegligibly small ones. We develop Conformal Risk Adaptation (CRA), which\nintroduces a new score function for creating adaptive prediction sets that\nsignificantly improve conditional risk control for segmentation tasks. We\nestablish a novel theoretical framework that demonstrates a fundamental\nconnection between conformal risk control and conformal prediction through a\nweighted quantile approach, applicable to any score function. To address the\nchallenge of poorly calibrated probabilities in segmentation models, we\nintroduce a specialized probability calibration framework that enhances the\nreliability of pixel-wise inclusion estimates. Using these calibrated\nprobabilities, we propose Calibrated Conformal Risk Adaptation (CCRA) and a\nstratified variant (CCRA-S) that partitions images based on their\ncharacteristics and applies group-specific thresholds to further enhance\nconditional risk control. Our experiments on polyp segmentation demonstrate\nthat all three methods (CRA, CCRA, and CCRA-S) provide valid marginal risk\ncontrol and deliver more consistent conditional risk control across diverse\nimages compared to standard approaches, offering a principled approach to\nuncertainty quantification that is particularly valuable for high-stakes and\npersonalized segmentation applications.\n", "date": "2025-04-10", "categories": []} {"id": "2406.00539", "title": "CONFINE: Conformal Prediction for Interpretable Neural Networks", "abstract": " Deep neural networks exhibit remarkable performance, yet their black-box\nnature limits their utility in fields like healthcare where interpretability is\ncrucial. Existing explainability approaches often sacrifice accuracy and lack\nquantifiable measures of prediction uncertainty. In this study, we introduce\nConformal Prediction for Interpretable Neural Networks (CONFINE), a versatile\nframework that generates prediction sets with statistically robust uncertainty\nestimates instead of point predictions to enhance model transparency and\nreliability. CONFINE not only provides example-based explanations and\nconfidence estimates for individual predictions but also boosts accuracy by up\nto 3.6%. We define a new metric, correct efficiency, to evaluate the fraction\nof prediction sets that contain precisely the correct label and show that\nCONFINE achieves correct efficiency of up to 3.3% higher than the original\naccuracy, matching or exceeding prior methods. CONFINE's marginal and\nclass-conditional coverages attest to its validity across tasks spanning\nmedical image classification to language understanding. Being adaptable to any\npre-trained classifier, CONFINE marks a significant advance towards transparent\nand trustworthy deep learning applications in critical domains.\n", "date": "2024-06-01", "categories": []} {"id": "2504.02919", "title": "ConfEviSurrogate: A Conformalized Evidential Surrogate Model for\n Uncertainty Quantification", "abstract": " Surrogate models, crucial for approximating complex simulation data across\nsciences, inherently carry uncertainties that range from simulation noise to\nmodel prediction errors. Without rigorous uncertainty quantification,\npredictions become unreliable and hence hinder analysis. While methods like\nMonte Carlo dropout and ensemble models exist, they are often costly, fail to\nisolate uncertainty types, and lack guaranteed coverage in prediction\nintervals. To address this, we introduce ConfEviSurrogate, a novel\nConformalized Evidential Surrogate Model that can efficiently learn high-order\nevidential distributions, directly predict simulation outcomes, separate\nuncertainty sources, and provide prediction intervals. A conformal\nprediction-based calibration step further enhances interval reliability to\nensure coverage and improve efficiency. Our ConfEviSurrogate demonstrates\naccurate predictions and robust uncertainty estimates in diverse simulations,\nincluding cosmology, ocean dynamics, and fluid dynamics.\n", "date": "2025-04-03", "categories": []} {"id": "2403.16871", "title": "Conformal Off-Policy Prediction for Multi-Agent Systems", "abstract": " Off-Policy Prediction (OPP), i.e., predicting the outcomes of a target policy\nusing only data collected under a nominal (behavioural) policy, is a paramount\nproblem in data-driven analysis of safety-critical systems where the deployment\nof a new policy may be unsafe. To achieve dependable off-policy predictions,\nrecent work on Conformal Off-Policy Prediction (COPP) leverage the conformal\nprediction framework to derive prediction regions with probabilistic guarantees\nunder the target process. Existing COPP methods can account for the\ndistribution shifts induced by policy switching, but are limited to\nsingle-agent systems and scalar outcomes (e.g., rewards). In this work, we\nintroduce MA-COPP, the first conformal prediction method to solve OPP problems\ninvolving multi-agent systems, deriving joint prediction regions for all\nagents' trajectories when one or more ego agents change their policies. Unlike\nthe single-agent scenario, this setting introduces higher complexity as the\ndistribution shifts affect predictions for all agents, not just the ego agents,\nand the prediction task involves full multi-dimensional trajectories, not just\nreward values. A key contribution of MA-COPP is to avoid enumeration or\nexhaustive search of the output space of agent trajectories, which is instead\nrequired by existing COPP methods to construct the prediction region. We\nachieve this by showing that an over-approximation of the true joint prediction\nregion (JPR) can be constructed, without enumeration, from the maximum density\nratio of the JPR trajectories. We evaluate the effectiveness of MA-COPP in\nmulti-agent systems from the PettingZoo library and the F1TENTH autonomous\nracing environment, achieving nominal coverage in higher dimensions and various\nshift settings.\n", "date": "2024-03-25", "categories": []} {"id": "2503.23819", "title": "Conformal uncertainty quantification to evaluate predictive fairness of\n foundation AI model for skin lesion classes across patient demographics", "abstract": " Deep learning based diagnostic AI systems based on medical images are\nstarting to provide similar performance as human experts. However these data\nhungry complex systems are inherently black boxes and therefore slow to be\nadopted for high risk applications like healthcare. This problem of lack of\ntransparency is exacerbated in the case of recent large foundation models,\nwhich are trained in a self supervised manner on millions of data points to\nprovide robust generalisation across a range of downstream tasks, but the\nembeddings generated from them happen through a process that is not\ninterpretable, and hence not easily trustable for clinical applications. To\naddress this timely issue, we deploy conformal analysis to quantify the\npredictive uncertainty of a vision transformer (ViT) based foundation model\nacross patient demographics with respect to sex, age and ethnicity for the\ntasks of skin lesion classification using several public benchmark datasets.\nThe significant advantage of this method is that conformal analysis is method\nindependent and it not only provides a coverage guarantee at population level\nbut also provides an uncertainty score for each individual. We used a\nmodel-agnostic dynamic F1-score-based sampling during model training, which\nhelped to stabilize the class imbalance and we investigate the effects on\nuncertainty quantification (UQ) with or without this bias mitigation step. Thus\nwe show how this can be used as a fairness metric to evaluate the robustness of\nthe feature embeddings of the foundation model (Google DermFoundation) and thus\nadvance the trustworthiness and fairness of clinical AI.\n", "date": "2025-03-31", "categories": []} {"id": "2406.17819", "title": "Automatically Adaptive Conformal Risk Control", "abstract": " Science and technology have a growing need for effective mechanisms that\nensure reliable, controlled performance from black-box machine learning\nalgorithms. These performance guarantees should ideally hold conditionally on\nthe input-that is the performance guarantees should hold, at least\napproximately, no matter what the input. However, beyond stylized discrete\ngroupings such as ethnicity and gender, the right notion of conditioning can be\ndifficult to define. For example, in problems such as image segmentation, we\nwant the uncertainty to reflect the intrinsic difficulty of the test sample,\nbut this may be difficult to capture via a conditioning event. Building on the\nrecent work of Gibbs et al. [2023], we propose a methodology for achieving\napproximate conditional control of statistical risks-the expected value of loss\nfunctions-by adapting to the difficulty of test samples. Our framework goes\nbeyond traditional conditional risk control based on user-provided conditioning\nevents to the algorithmic, data-driven determination of appropriate function\nclasses for conditioning. We apply this framework to various regression and\nsegmentation tasks, enabling finer-grained control over model performance and\ndemonstrating that by continuously monitoring and adjusting these parameters,\nwe can achieve superior precision compared to conventional risk-control\nmethods.\n", "date": "2024-06-25", "categories": []} {"id": "2405.19230", "title": "Valid Conformal Prediction for Dynamic GNNs", "abstract": " Dynamic graphs provide a flexible data abstraction for modelling many sorts\nof real-world systems, such as transport, trade, and social networks. Graph\nneural networks (GNNs) are powerful tools allowing for different kinds of\nprediction and inference on these systems, but getting a handle on uncertainty,\nespecially in dynamic settings, is a challenging problem. In this work we\npropose to use a dynamic graph representation known in the tensor literature as\nthe unfolding, to achieve valid prediction sets via conformal prediction. This\nrepresentation, a simple graph, can be input to any standard GNN and does not\nrequire any modification to existing GNN architectures or conformal prediction\nroutines. One of our key contributions is a careful mathematical consideration\nof the different inference scenarios which can arise in a dynamic graph\nmodelling context. For a range of practically relevant cases, we obtain valid\nprediction sets with almost no assumptions, even dispensing with\nexchangeability. In a more challenging scenario, which we call the\nsemi-inductive regime, we achieve valid prediction under stronger assumptions,\nakin to stationarity. We provide real data examples demonstrating validity,\nshowing improved accuracy over baselines, and sign-posting different failure\nmodes which can occur when those assumptions are violated.\n", "date": "2024-05-29", "categories": []} {"id": "2405.11968", "title": "Conditional Shift-Robust Conformal Prediction for Graph Neural Network", "abstract": " Graph Neural Networks (GNNs) have emerged as potent tools for predicting\noutcomes in graph-structured data. Despite their efficacy, a significant\ndrawback of GNNs lies in their limited ability to provide robust uncertainty\nestimates, posing challenges to their reliability in contexts where errors\ncarry significant consequences. Moreover, GNNs typically excel in\nin-distribution settings, assuming that training and test data follow identical\ndistributions a condition often unmet in real world graph data scenarios. In\nthis article, we leverage conformal prediction, a widely recognized statistical\ntechnique for quantifying uncertainty by transforming predictive model outputs\ninto prediction sets, to address uncertainty quantification in GNN predictions\namidst conditional shift\\footnote{Representing the change in conditional\nprobability distribution \\(P(label|input)\\) from source domain to target\ndomain.} in graph-based semi-supervised learning (SSL). Additionally, we\npropose a novel loss function aimed at refining model predictions by minimizing\nconditional shift in latent stages. Termed Conditional Shift Robust (CondSR)\nconformal prediction for GNNs, our approach CondSR is model-agnostic and\nadaptable to various classification models. We validate the effectiveness of\nour method on standard graph benchmark datasets, integrating it with\nstate-of-the-art GNNs in node classification tasks. Comprehensive evaluations\ndemonstrate that our approach consistently achieves any predefined target\nmarginal coverage, enhances the accuracy of state of the art GNN models by up\nto 12\\% under conditional shift, and reduces the prediction set size by up to\n48\\%. The code implementation is publicly available for further exploration and\nexperimentation.\n", "date": "2024-05-20", "categories": []} {"id": "2410.20579", "title": "Toward Conditional Distribution Calibration in Survival Prediction", "abstract": " Survival prediction often involves estimating the time-to-event distribution\nfrom censored datasets. Previous approaches have focused on enhancing\ndiscrimination and marginal calibration. In this paper, we highlight the\nsignificance of conditional calibration for real-world applications --\nespecially its role in individual decision-making. We propose a method based on\nconformal prediction that uses the model's predicted individual survival\nprobability at that instance's observed time. This method effectively improves\nthe model's marginal and conditional calibration, without compromising\ndiscrimination. We provide asymptotic theoretical guarantees for both marginal\nand conditional calibration and test it extensively across 15 diverse\nreal-world datasets, demonstrating the method's practical effectiveness and\nversatility in various settings.\n", "date": "2024-10-27", "categories": []} {"id": "2503.16809", "title": "Online Selective Conformal Prediction: Errors and Solutions", "abstract": " In online selective conformal inference, data arrives sequentially, and\nprediction intervals are constructed only when an online selection rule is met.\nSince online selections may break the exchangeability between the selected test\ndatum and the rest of the data, one must correct for this by suitably selecting\nthe calibration data. In this paper, we evaluate existing calibration selection\nstrategies and pinpoint some fundamental errors in the associated claims that\nguarantee selection-conditional coverage and control of the false coverage rate\n(FCR). To address these shortcomings, we propose novel calibration selection\nstrategies that provably preserve the exchangeability of the calibration data\nand the selected test datum. Consequently, we demonstrate that online selective\nconformal inference with these strategies guarantees both selection-conditional\ncoverage and FCR control. Our theoretical findings are supported by\nexperimental evidence examining tradeoffs between valid methods.\n", "date": "2025-03-21", "categories": []} {"id": "2503.14663", "title": "Sepsyn-OLCP: An Online Learning-based Framework for Early Sepsis\n Prediction with Uncertainty Quantification using Conformal Prediction", "abstract": " Sepsis is a life-threatening syndrome with high morbidity and mortality in\nhospitals. Early prediction of sepsis plays a crucial role in facilitating\nearly interventions for septic patients. However, early sepsis prediction\nsystems with uncertainty quantification and adaptive learning are scarce. This\npaper proposes Sepsyn-OLCP, a novel online learning algorithm for early sepsis\nprediction by integrating conformal prediction for uncertainty quantification\nand Bayesian bandits for adaptive decision-making. By combining the robustness\nof Bayesian models with the statistical uncertainty guarantees of conformal\nprediction methodologies, this algorithm delivers accurate and trustworthy\npredictions, addressing the critical need for reliable and adaptive systems in\nhigh-stakes healthcare applications such as early sepsis prediction. We\nevaluate the performance of Sepsyn-OLCP in terms of regret in stochastic bandit\nsetting, the area under the receiver operating characteristic curve (AUROC),\nand F-measure. Our results show that Sepsyn-OLCP outperforms existing\nindividual models, increasing AUROC of a neural network from 0.64 to 0.73\nwithout retraining and high computational costs. And the model selection policy\nconverges to the optimal strategy in the long run. We propose a novel\nreinforcement learning-based framework integrated with conformal prediction\ntechniques to provide uncertainty quantification for early sepsis prediction.\nThe proposed methodology delivers accurate and trustworthy predictions,\naddressing a critical need in high-stakes healthcare applications like early\nsepsis prediction.\n", "date": "2025-03-18", "categories": []} {"id": "2410.09408", "title": "C-Adapter: Adapting Deep Classifiers for Efficient Conformal Prediction\n Sets", "abstract": " Conformal prediction, as an emerging uncertainty quantification technique,\ntypically functions as post-hoc processing for the outputs of trained\nclassifiers. To optimize the classifier for maximum predictive efficiency,\nConformal Training rectifies the training objective with a regularization that\nminimizes the average prediction set size at a specific error rate. However,\nthe regularization term inevitably deteriorates the classification accuracy and\nleads to suboptimal efficiency of conformal predictors. To address this issue,\nwe introduce \\textbf{Conformal Adapter} (C-Adapter), an adapter-based tuning\nmethod to enhance the efficiency of conformal predictors without sacrificing\naccuracy. In particular, we implement the adapter as a class of intra\norder-preserving functions and tune it with our proposed loss that maximizes\nthe discriminability of non-conformity scores between correctly and randomly\nmatched data-label pairs. Using C-Adapter, the model tends to produce extremely\nhigh non-conformity scores for incorrect labels, thereby enhancing the\nefficiency of prediction sets across different coverage rates. Extensive\nexperiments demonstrate that C-Adapter can effectively adapt various\nclassifiers for efficient prediction sets, as well as enhance the conformal\ntraining method.\n", "date": "2024-10-12", "categories": []} {"id": "2503.11709", "title": "Conformal Prediction and Human Decision Making", "abstract": " Methods to quantify uncertainty in predictions from arbitrary models are in\ndemand in high-stakes domains like medicine and finance. Conformal prediction\nhas emerged as a popular method for producing a set of predictions with\nspecified average coverage, in place of a single prediction and confidence\nvalue. However, the value of conformal prediction sets to assist human\ndecisions remains elusive due to the murky relationship between coverage\nguarantees and decision makers' goals and strategies. How should we think about\nconformal prediction sets as a form of decision support? We outline a decision\ntheoretic framework for evaluating predictive uncertainty as informative\nsignals, then contrast what can be said within this framework about idealized\nuse of calibrated probabilities versus conformal prediction sets. Informed by\nprior empirical results and theories of human decisions under uncertainty, we\nformalize a set of possible strategies by which a decision maker might use a\nprediction set. We identify ways in which conformal prediction sets and posthoc\npredictive uncertainty quantification more broadly are in tension with common\ngoals and needs in human-AI decision making. We give recommendations for future\nresearch in predictive uncertainty quantification to support human decision\nmakers.\n", "date": "2025-03-12", "categories": []} {"id": "2410.09878", "title": "Provably Reliable Conformal Prediction Sets in the Presence of Data\n Poisoning", "abstract": " Conformal prediction provides model-agnostic and distribution-free\nuncertainty quantification through prediction sets that are guaranteed to\ninclude the ground truth with any user-specified probability. Yet, conformal\nprediction is not reliable under poisoning attacks where adversaries manipulate\nboth training and calibration data, which can significantly alter prediction\nsets in practice. As a solution, we propose reliable prediction sets (RPS): the\nfirst efficient method for constructing conformal prediction sets with provable\nreliability guarantees under poisoning. To ensure reliability under training\npoisoning, we introduce smoothed score functions that reliably aggregate\npredictions of classifiers trained on distinct partitions of the training data.\nTo ensure reliability under calibration poisoning, we construct multiple\nprediction sets, each calibrated on distinct subsets of the calibration data.\nWe then aggregate them into a majority prediction set, which includes a class\nonly if it appears in a majority of the individual sets. Both proposed\naggregations mitigate the influence of datapoints in the training and\ncalibration data on the final prediction set. We experimentally validate our\napproach on image classification tasks, achieving strong reliability while\nmaintaining utility and preserving coverage on clean data. Overall, our\napproach represents an important step towards more trustworthy uncertainty\nquantification in the presence of data poisoning.\n", "date": "2024-10-13", "categories": []} {"id": "2503.04191", "title": "Conformal forecasting for surgical instrument trajectory", "abstract": " Forecasting surgical instrument trajectories and predicting the next surgical\naction recently started to attract attention from the research community. Both\nthese tasks are crucial for automation and assistance in endoscopy surgery.\nGiven the safety-critical nature of these tasks, reliable uncertainty\nquantification is essential. Conformal prediction is a fast-growing and widely\nrecognized framework for uncertainty estimation in machine learning and\ncomputer vision, offering distribution-free, theoretically valid prediction\nintervals. In this work, we explore the application of standard conformal\nprediction and conformalized quantile regression to estimate uncertainty in\nforecasting surgical instrument motion, i.e., predicting direction and\nmagnitude of surgical instruments' future motion. We analyze and compare their\ncoverage and interval sizes, assessing the impact of multiple hypothesis\ntesting and correction methods. Additionally, we show how these techniques can\nbe employed to produce useful uncertainty heatmaps. To the best of our\nknowledge, this is the first study applying conformal prediction to surgical\nguidance, marking an initial step toward constructing principled prediction\nintervals with formal coverage guarantees in this domain.\n", "date": "2025-03-06", "categories": []} {"id": "2410.06494", "title": "Conformal Prediction: A Data Perspective", "abstract": " Conformal prediction (CP), a distribution-free uncertainty quantification\n(UQ) framework, reliably provides valid predictive inference for black-box\nmodels. CP constructs prediction sets that contain the true output with a\nspecified probability. However, modern data science diverse modalities, along\nwith increasing data and model complexity, challenge traditional CP methods.\nThese developments have spurred novel approaches to address evolving scenarios.\nThis survey reviews the foundational concepts of CP and recent advancements\nfrom a data-centric perspective, including applications to structured,\nunstructured, and dynamic data. We also discuss the challenges and\nopportunities CP faces in large-scale data and models.\n", "date": "2024-10-09", "categories": []} {"id": "2411.02114", "title": "Semiparametric conformal prediction", "abstract": " Many risk-sensitive applications require well-calibrated prediction sets over\nmultiple, potentially correlated target variables, for which the prediction\nalgorithm may report correlated errors. In this work, we aim to construct the\nconformal prediction set accounting for the joint correlation structure of the\nvector-valued non-conformity scores. Drawing from the rich literature on\nmultivariate quantiles and semiparametric statistics, we propose an algorithm\nto estimate the $1-\\alpha$ quantile of the scores, where $\\alpha$ is the\nuser-specified miscoverage rate. In particular, we flexibly estimate the joint\ncumulative distribution function (CDF) of the scores using nonparametric vine\ncopulas and improve the asymptotic efficiency of the quantile estimate using\nits influence function. The vine decomposition allows our method to scale well\nto a large number of targets. As well as guaranteeing asymptotically exact\ncoverage, our method yields desired coverage and competitive efficiency on a\nrange of real-world regression problems, including those with missing-at-random\nlabels in the calibration set.\n", "date": "2024-11-04", "categories": []} {"id": "2410.06296", "title": "Conformal Structured Prediction", "abstract": " Conformal prediction has recently emerged as a promising strategy for\nquantifying the uncertainty of a predictive model; these algorithms modify the\nmodel to output sets of labels that are guaranteed to contain the true label\nwith high probability. However, existing conformal prediction algorithms have\nlargely targeted classification and regression settings, where the structure of\nthe prediction set has a simple form as a level set of the scoring function.\nHowever, for complex structured outputs such as text generation, these\nprediction sets might include a large number of labels and therefore be hard\nfor users to interpret. In this paper, we propose a general framework for\nconformal prediction in the structured prediction setting, that modifies\nexisting conformal prediction algorithms to output structured prediction sets\nthat implicitly represent sets of labels. In addition, we demonstrate how our\napproach can be applied in domains where the prediction sets can be represented\nas a set of nodes in a directed acyclic graph; for instance, for hierarchical\nlabels such as image classification, a prediction set might be a small subset\nof coarse labels implicitly representing the prediction set of all their more\nfine-descendants. We demonstrate how our algorithm can be used to construct\nprediction sets that satisfy a desired coverage guarantee in several domains.\n", "date": "2024-10-08", "categories": []} {"id": "2503.05239", "title": "Robust Conformal Prediction with a Single Binary Certificate", "abstract": " Conformal prediction (CP) converts any model's output to prediction sets with\na guarantee to cover the true label with (adjustable) high probability. Robust\nCP extends this guarantee to worst-case (adversarial) inputs. Existing\nbaselines achieve robustness by bounding randomly smoothed conformity scores.\nIn practice, they need expensive Monte-Carlo (MC) sampling (e.g. $\\sim10^4$\nsamples per point) to maintain an acceptable set size. We propose a robust\nconformal prediction that produces smaller sets even with significantly lower\nMC samples (e.g. 150 for CIFAR10). Our approach binarizes samples with an\nadjustable (or automatically adjusted) threshold selected to preserve the\ncoverage guarantee. Remarkably, we prove that robustness can be achieved by\ncomputing only one binary certificate, unlike previous methods that certify\neach calibration (or test) point. Thus, our method is faster and returns\nsmaller robust sets. We also eliminate a previous limitation that requires a\nbounded score function.\n", "date": "2025-03-07", "categories": []} {"id": "2503.05618", "title": "Conformal Prediction for Image Segmentation Using Morphological\n Prediction Sets", "abstract": " Image segmentation is a challenging task influenced by multiple sources of\nuncertainty, such as the data labeling process or the sampling of training\ndata. In this paper we focus on binary segmentation and address these\nchallenges using conformal prediction, a family of model- and data-agnostic\nmethods for uncertainty quantification that provide finite-sample theoretical\nguarantees and applicable to any pretrained predictor. Our approach involves\ncomputing nonconformity scores, a type of prediction residual, on held-out\ncalibration data not used during training. We use dilation, one of the\nfundamental operations in mathematical morphology, to construct a margin added\nto the borders of predicted segmentation masks. At inference, the predicted set\nformed by the mask and its margin contains the ground-truth mask with high\nprobability, at a confidence level specified by the user. The size of the\nmargin serves as an indicator of predictive uncertainty for a given model and\ndataset. We work in a regime of minimal information as we do not require any\nfeedback from the predictor: only the predicted masks are needed for computing\nthe prediction sets. Hence, our method is applicable to any segmentation model,\nincluding those based on deep learning; we evaluate our approach on several\nmedical imaging applications.\n", "date": "2025-03-07", "categories": []} {"id": "2501.13430", "title": "Wasserstein-regularized Conformal Prediction under General Distribution\n Shift", "abstract": " Conformal prediction yields a prediction set with guaranteed $1-\\alpha$\ncoverage of the true target under the i.i.d. assumption, which may not hold and\nlead to a gap between $1-\\alpha$ and the actual coverage. Prior studies bound\nthe gap using total variation distance, which cannot identify the gap changes\nunder distribution shift at a given $\\alpha$. Besides, existing methods are\nmostly limited to covariate shift,while general joint distribution shifts are\nmore common in practice but less researched.In response, we first propose a\nWasserstein distance-based upper bound of the coverage gap and analyze the\nbound using probability measure pushforwards between the shifted joint data and\nconformal score distributions, enabling a separation of the effect of covariate\nand concept shifts over the coverage gap. We exploit the separation to design\nan algorithm based on importance weighting and regularized representation\nlearning (WR-CP) to reduce the Wasserstein bound with a finite-sample error\nbound.WR-CP achieves a controllable balance between conformal prediction\naccuracy and efficiency. Experiments on six datasets prove that WR-CP can\nreduce coverage gaps to $3.2\\%$ across different confidence levels and outputs\nprediction sets 37$\\%$ smaller than the worst-case approach on average.\n", "date": "2025-01-23", "categories": []} {"id": "2312.07434", "title": "Multi-Modal Conformal Prediction Regions with Simple Structures by\n Optimizing Convex Shape Templates", "abstract": " Conformal prediction is a statistical tool for producing prediction regions\nfor machine learning models that are valid with high probability. A key\ncomponent of conformal prediction algorithms is a \\emph{non-conformity score\nfunction} that quantifies how different a model's prediction is from the\nunknown ground truth value. Essentially, these functions determine the shape\nand the size of the conformal prediction regions. While prior work has gone\ninto creating score functions that produce multi-model prediction regions, such\nregions are generally too complex for use in downstream planning and control\nproblems. We propose a method that optimizes parameterized \\emph{shape template\nfunctions} over calibration data, which results in non-conformity score\nfunctions that produce prediction regions with minimum volume. Our approach\nresults in prediction regions that are \\emph{multi-modal}, so they can properly\ncapture residuals of distributions that have multiple modes, and\n\\emph{practical}, so each region is convex and can be easily incorporated into\ndownstream tasks, such as a motion planner using conformal prediction regions.\nOur method applies to general supervised learning tasks, while we illustrate\nits use in time-series prediction. We provide a toolbox and present\nillustrative case studies of F16 fighter jets and autonomous vehicles, showing\nan up to $68\\%$ reduction in prediction region area compared to a circular\nbaseline region.\n", "date": "2023-12-12", "categories": []} {"id": "2407.10230", "title": "Weighted Aggregation of Conformity Scores for Classification", "abstract": " Conformal prediction is a powerful framework for constructing prediction sets\nwith valid coverage guarantees in multi-class classification. However, existing\nmethods often rely on a single score function, which can limit their efficiency\nand informativeness. We propose a novel approach that combines multiple score\nfunctions to improve the performance of conformal predictors by identifying\noptimal weights that minimize prediction set size. Our theoretical analysis\nestablishes a connection between the weighted score functions and subgraph\nclasses of functions studied in Vapnik-Chervonenkis theory, providing a\nrigorous mathematical basis for understanding the effectiveness of the proposed\nmethod. Experiments demonstrate that our approach consistently outperforms\nsingle-score conformal predictors while maintaining valid coverage, offering a\nprincipled and data-driven way to enhance the efficiency and practicality of\nconformal prediction in classification tasks.\n", "date": "2024-07-14", "categories": []} {"id": "2503.02205", "title": "Volume-Sorted Prediction Set: Efficient Conformal Prediction for\n Multi-Target Regression", "abstract": " We introduce Volume-Sorted Prediction Set (VSPS), a novel method for\nuncertainty quantification in multi-target regression that uses conditional\nnormalizing flows with conformal calibration. This approach constructs\nflexible, non-convex predictive regions with guaranteed coverage probabilities,\novercoming limitations of traditional methods. By learning a transformation\nwhere the conditional distribution of responses follows a known form, VSPS\nidentifies dense regions in the original space using the Jacobian determinant.\nThis enables the creation of prediction regions that adapt to the true\nunderlying distribution, focusing on areas of high probability density.\nExperimental results demonstrate that VSPS produces smaller, more informative\nprediction regions while maintaining robust coverage guarantees, enhancing\nuncertainty modeling in complex, high-dimensional settings.\n", "date": "2025-03-04", "categories": []} {"id": "2410.01767", "title": "Utility-Directed Conformal Prediction: A Decision-Aware Framework for\n Actionable Uncertainty Quantification", "abstract": " Interest has been growing in decision-focused machine learning methods which\ntrain models to account for how their predictions are used in downstream\noptimization problems. Doing so can often improve performance on subsequent\ndecision problems. However, current methods for uncertainty quantification do\nnot incorporate any information about downstream decisions. We develop a\nmethodology based on conformal prediction to identify prediction sets that\naccount for a downstream cost function, making them more appropriate to inform\nhigh-stakes decision-making. Our approach harnesses the strengths of conformal\nmethods -- modularity, model-agnosticism, and statistical coverage guarantees\n-- while incorporating downstream decisions and user-specified utility\nfunctions. We prove that our methods retain standard coverage guarantees.\nEmpirical evaluation across a range of datasets and utility metrics\ndemonstrates that our methods achieve significantly lower costs than standard\nconformal methods. We present a real-world use case in healthcare diagnosis,\nwhere our method effectively incorporates the hierarchical structure of\ndermatological diseases. The method successfully generates sets with coherent\ndiagnostic meaning, potentially aiding triage for dermatology diagnosis and\nillustrating how our method can ground high-stakes decision-making employing\ndomain knowledge.\n", "date": "2024-10-02", "categories": []} {"id": "2502.17744", "title": "Conformal Prediction Under Generalized Covariate Shift with Posterior\n Drift", "abstract": " In many real applications of statistical learning, collecting sufficiently\nmany training data is often expensive, time-consuming, or even unrealistic. In\nthis case, a transfer learning approach, which aims to leverage knowledge from\na related source domain to improve the learning performance in the target\ndomain, is more beneficial. There have been many transfer learning methods\ndeveloped under various distributional assumptions. In this article, we study a\nparticular type of classification problem, called conformal prediction, under a\nnew distributional assumption for transfer learning. Classifiers under the\nconformal prediction framework predict a set of plausible labels instead of one\nsingle label for each data instance, affording a more cautious and safer\ndecision. We consider a generalization of the \\textit{covariate shift with\nposterior drift} setting for transfer learning. Under this setting, we propose\na weighted conformal classifier that leverages both the source and target\nsamples, with a coverage guarantee in the target domain. Theoretical studies\ndemonstrate favorable asymptotic properties. Numerical studies further\nillustrate the usefulness of the proposed method.\n", "date": "2025-02-25", "categories": []} {"id": "2409.07902", "title": "Conformal Distributed Remote Inference in Sensor Networks Under\n Reliability and Communication Constraints", "abstract": " This paper presents communication-constrained distributed conformal risk\ncontrol (CD-CRC) framework, a novel decision-making framework for sensor\nnetworks under communication constraints. Targeting multi-label classification\nproblems, such as segmentation, CD-CRC dynamically adjusts local and global\nthresholds used to identify significant labels with the goal of ensuring a\ntarget false negative rate (FNR), while adhering to communication capacity\nlimits. CD-CRC builds on online exponentiated gradient descent to estimate the\nrelative quality of the observations of different sensors, and on online\nconformal risk control (CRC) as a mechanism to control local and global\nthresholds. CD-CRC is proved to offer deterministic worst-case performance\nguarantees in terms of FNR and communication overhead, while the regret\nperformance in terms of false positive rate (FPR) is characterized as a\nfunction of the key hyperparameters. Simulation results highlight the\neffectiveness of CD-CRC, particularly in communication resource-constrained\nenvironments, making it a valuable tool for enhancing the performance and\nreliability of distributed sensor networks.\n", "date": "2024-09-12", "categories": []} {"id": "2502.14773", "title": "Sparse Activations as Conformal Predictors", "abstract": " Conformal prediction is a distribution-free framework for uncertainty\nquantification that replaces point predictions with sets, offering marginal\ncoverage guarantees (i.e., ensuring that the prediction sets contain the true\nlabel with a specified probability, in expectation). In this paper, we uncover\na novel connection between conformal prediction and sparse softmax-like\ntransformations, such as sparsemax and $\\gamma$-entmax (with $\\gamma > 1$),\nwhich may assign nonzero probability only to a subset of labels. We introduce\nnew non-conformity scores for classification that make the calibration process\ncorrespond to the widely used temperature scaling method. At test time,\napplying these sparse transformations with the calibrated temperature leads to\na support set (i.e., the set of labels with nonzero probability) that\nautomatically inherits the coverage guarantees of conformal prediction. Through\nexperiments on computer vision and text classification benchmarks, we\ndemonstrate that the proposed method achieves competitive results in terms of\ncoverage, efficiency, and adaptiveness compared to standard non-conformity\nscores based on softmax.\n", "date": "2025-02-20", "categories": []} {"id": "2502.16658", "title": "Volume Optimality in Conformal Prediction with Structured Prediction\n Sets", "abstract": " Conformal Prediction is a widely studied technique to construct prediction\nsets of future observations. Most conformal prediction methods focus on\nachieving the necessary coverage guarantees, but do not provide formal\nguarantees on the size (volume) of the prediction sets. We first prove an\nimpossibility of volume optimality where any distribution-free method can only\nfind a trivial solution. We then introduce a new notion of volume optimality by\nrestricting the prediction sets to belong to a set family (of finite\nVC-dimension), specifically a union of $k$-intervals. Our main contribution is\nan efficient distribution-free algorithm based on dynamic programming (DP) to\nfind a union of $k$-intervals that is guaranteed for any distribution to have\nnear-optimal volume among all unions of $k$-intervals satisfying the desired\ncoverage property. By adopting the framework of distributional conformal\nprediction (Chernozhukov et al., 2021), the new DP based conformity score can\nalso be applied to achieve approximate conditional coverage and conditional\nrestricted volume optimality, as long as a reasonable estimator of the\nconditional CDF is available. While the theoretical results already establish\nvolume-optimality guarantees, they are complemented by experiments that\ndemonstrate that our method can significantly outperform existing methods in\nmany settings.\n", "date": "2025-02-23", "categories": []} {"id": "2405.02140", "title": "An Information Theoretic Perspective on Conformal Prediction", "abstract": " Conformal Prediction (CP) is a distribution-free uncertainty estimation\nframework that constructs prediction sets guaranteed to contain the true answer\nwith a user-specified probability. Intuitively, the size of the prediction set\nencodes a general notion of uncertainty, with larger sets associated with\nhigher degrees of uncertainty. In this work, we leverage information theory to\nconnect conformal prediction to other notions of uncertainty. More precisely,\nwe prove three different ways to upper bound the intrinsic uncertainty, as\ndescribed by the conditional entropy of the target variable given the inputs,\nby combining CP with information theoretical inequalities. Moreover, we\ndemonstrate two direct and useful applications of such connection between\nconformal prediction and information theory: (i) more principled and effective\nconformal training objectives that generalize previous approaches and enable\nend-to-end training of machine learning models from scratch, and (ii) a natural\nmechanism to incorporate side information into conformal prediction. We\nempirically validate both applications in centralized and federated learning\nsettings, showing our theoretical results translate to lower inefficiency\n(average prediction set size) for popular CP methods.\n", "date": "2024-05-03", "categories": []} {"id": "2410.01660", "title": "Conformal Generative Modeling with Improved Sample Efficiency through\n Sequential Greedy Filtering", "abstract": " Generative models lack rigorous statistical guarantees for their outputs and\nare therefore unreliable in safety-critical applications. In this work, we\npropose Sequential Conformal Prediction for Generative Models (SCOPE-Gen), a\nsequential conformal prediction method producing prediction sets that satisfy a\nrigorous statistical guarantee called conformal admissibility control. This\nguarantee states that with high probability, the prediction sets contain at\nleast one admissible (or valid) example. To this end, our method first samples\nan initial set of i.i.d. examples from a black box generative model. Then, this\nset is iteratively pruned via so-called greedy filters. As a consequence of the\niterative generation procedure, admissibility of the final prediction set\nfactorizes as a Markov chain. This factorization is crucial, because it allows\nto control each factor separately, using conformal prediction. In comparison to\nprior work, our method demonstrates a large reduction in the number of\nadmissibility evaluations during calibration. This reduction is important in\nsafety-critical applications, where these evaluations must be conducted\nmanually by domain experts and are therefore costly and time consuming. We\nhighlight the advantages of our method in terms of admissibility evaluations\nand cardinality of the prediction sets through experiments in natural language\ngeneration and molecular graph extension tasks.\n", "date": "2024-10-02", "categories": []} {"id": "2412.01098", "title": "Spatial Conformal Inference through Localized Quantile Regression", "abstract": " Reliable uncertainty quantification at unobserved spatial locations,\nespecially in the presence of complex and heterogeneous datasets, remains a\ncore challenge in spatial statistics. Traditional approaches like Kriging rely\nheavily on assumptions such as normality, which often break down in\nlarge-scale, diverse datasets, leading to unreliable prediction intervals.\nWhile machine learning methods have emerged as powerful alternatives, they\nprimarily focus on point predictions and provide limited mechanisms for\nuncertainty quantification. Conformal prediction, a distribution-free\nframework, offers valid prediction intervals without relying on parametric\nassumptions. However, existing conformal prediction methods are either not\ntailored for spatial settings, or existing ones for spatial data have relied on\nrather restrictive i.i.d. assumptions. In this paper, we propose Localized\nSpatial Conformal Prediction (LSCP), a conformal prediction method designed\nspecifically for spatial data. LSCP leverages localized quantile regression to\nconstruct prediction intervals. Instead of i.i.d. assumptions, our theoretical\nanalysis builds on weaker conditions of stationarity and spatial mixing, which\nis natural for spatial data, providing finite-sample bounds on the conditional\ncoverage gap and establishing asymptotic guarantees for conditional coverage.\nWe present experiments on both synthetic and real-world datasets to demonstrate\nthat LSCP achieves accurate coverage with significantly tighter and more\nconsistent prediction intervals across the spatial domain compared to existing\nmethods.\n", "date": "2024-12-02", "categories": []} {"id": "2502.10947", "title": "The Relationship between No-Regret Learning and Online Conformal\n Prediction", "abstract": " Existing algorithms for online conformal prediction -- guaranteeing marginal\ncoverage in adversarial settings -- are variants of online gradient descent\n(OGD), but their analyses of worst-case coverage do not follow from the regret\nguarantee of OGD. What is the relationship between no-regret learning and\nonline conformal prediction? We observe that although standard regret\nguarantees imply marginal coverage in i.i.d. settings, this connection fails as\nsoon as we either move to adversarial environments or ask for group conditional\ncoverage. On the other hand, we show a tight connection between threshold\ncalibrated coverage and swap-regret in adversarial settings, which extends to\ngroup-conditional (multi-valid) coverage. We also show that algorithms in the\nfollow the perturbed leader family of no regret learning algorithms (which\nincludes online gradient descent) can be used to give group-conditional\ncoverage guarantees in adversarial settings for arbitrary grouping functions.\nVia this connection we analyze and conduct experiments using a multi-group\ngeneralization of the ACI algorithm of Gibbs & Candes [2021]\n(arXiv:2106.00170).\n", "date": "2025-02-16", "categories": []} {"id": "2410.01888", "title": "Conformal Prediction Sets Can Cause Disparate Impact", "abstract": " Conformal prediction is a statistically rigorous method for quantifying\nuncertainty in models by having them output sets of predictions, with larger\nsets indicating more uncertainty. However, prediction sets are not inherently\nactionable; many applications require a single output to act on, not several.\nTo overcome this limitation, prediction sets can be provided to a human who\nthen makes an informed decision. In any such system it is crucial to ensure the\nfairness of outcomes across protected groups, and researchers have proposed\nthat Equalized Coverage be used as the standard for fairness. By conducting\nexperiments with human participants, we demonstrate that providing prediction\nsets can lead to disparate impact in decisions. Disquietingly, we find that\nproviding sets that satisfy Equalized Coverage actually increases disparate\nimpact compared to marginal coverage. Instead of equalizing coverage, we\npropose to equalize set sizes across groups which empirically leads to lower\ndisparate impact.\n", "date": "2024-10-02", "categories": []} {"id": "2502.09985", "title": "On Volume Minimization in Conformal Regression", "abstract": " We study the question of volume optimality in split conformal regression, a\ntopic still poorly understood in comparison to coverage control. Using the fact\nthat the calibration step can be seen as an empirical volume minimization\nproblem, we first derive a finite-sample upper-bound on the excess volume loss\nof the interval returned by the classical split method. This important quantity\nmeasures the difference in length between the interval obtained with the split\nmethod and the shortest oracle prediction interval. Then, we introduce EffOrt,\na methodology that modifies the learning step so that the base prediction\nfunction is selected in order to minimize the length of the returned intervals.\nIn particular, our theoretical analysis of the excess volume loss of the\nprediction sets produced by EffOrt reveals the links between the learning and\ncalibration steps, and notably the impact of the choice of the function class\nof the base predictor. We also introduce Ad-EffOrt, an extension of the\nprevious method, which produces intervals whose size adapts to the value of the\ncovariate. Finally, we evaluate the empirical performance and the robustness of\nour methodologies.\n", "date": "2025-02-14", "categories": []} {"id": "2410.13735", "title": "Generative Conformal Prediction with Vectorized Non-Conformity Scores", "abstract": " Conformal prediction (CP) provides model-agnostic uncertainty quantification\nwith guaranteed coverage, but conventional methods often produce overly\nconservative uncertainty sets, especially in multi-dimensional settings. This\nlimitation arises from simplistic non-conformity scores that rely solely on\nprediction error, failing to capture the prediction error distribution's\ncomplexity. To address this, we propose a generative conformal prediction\nframework with vectorized non-conformity scores, leveraging a generative model\nto sample multiple predictions from the fitted data distribution. By computing\nnon-conformity scores across these samples and estimating empirical quantiles\nat different density levels, we construct adaptive uncertainty sets using\ndensity-ranked uncertainty balls. This approach enables more precise\nuncertainty allocation -- yielding larger prediction sets in high-confidence\nregions and smaller or excluded sets in low-confidence regions -- enhancing\nboth flexibility and efficiency. We establish theoretical guarantees for\nstatistical validity and demonstrate through extensive numerical experiments\nthat our method outperforms state-of-the-art techniques on synthetic and\nreal-world datasets.\n", "date": "2024-10-17", "categories": []} {"id": "2502.06884", "title": "Learning Conformal Abstention Policies for Adaptive Risk Management in\n Large Language and Vision-Language Models", "abstract": " Large Language and Vision-Language Models (LLMs/VLMs) are increasingly used\nin safety-critical applications, yet their opaque decision-making complicates\nrisk assessment and reliability. Uncertainty quantification (UQ) helps assess\nprediction confidence and enables abstention when uncertainty is high.\nConformal prediction (CP), a leading UQ method, provides statistical guarantees\nbut relies on static thresholds, which fail to adapt to task complexity and\nevolving data distributions, leading to suboptimal trade-offs in accuracy,\ncoverage, and informativeness. To address this, we propose learnable conformal\nabstention, integrating reinforcement learning (RL) with CP to optimize\nabstention thresholds dynamically. By treating CP thresholds as adaptive\nactions, our approach balances multiple objectives, minimizing prediction set\nsize while maintaining reliable coverage. Extensive evaluations across diverse\nLLM/VLM benchmarks show our method outperforms Least Ambiguous Classifiers\n(LAC) and Adaptive Prediction Sets (APS), improving accuracy by up to 3.2%,\nboosting AUROC for hallucination detection by 22.19%, enhancing\nuncertainty-guided selective generation (AUARC) by 21.17%, and reducing\ncalibration error by 70%-85%. These improvements hold across multiple models\nand datasets while consistently meeting the 90% coverage target, establishing\nour approach as a more effective and flexible solution for reliable\ndecision-making in safety-critical applications. The code is available at:\n{https://github.com/sinatayebati/vlm-uncertainty}.\n", "date": "2025-02-08", "categories": []} {"id": "2411.01696", "title": "Conformal Risk Minimization with Variance Reduction", "abstract": " Conformal prediction (CP) is a distribution-free framework for achieving\nprobabilistic guarantees on black-box models. CP is generally applied to a\nmodel post-training. Recent research efforts, on the other hand, have focused\non optimizing CP efficiency during training. We formalize this concept as the\nproblem of conformal risk minimization (CRM). In this direction, conformal\ntraining (ConfTr) by Stutz et al.(2022) is a technique that seeks to minimize\nthe expected prediction set size of a model by simulating CP in-between\ntraining updates. Despite its potential, we identify a strong source of sample\ninefficiency in ConfTr that leads to overly noisy estimated gradients,\nintroducing training instability and limiting practical use. To address this\nchallenge, we propose variance-reduced conformal training (VR-ConfTr), a CRM\nmethod that incorporates a variance reduction technique in the gradient\nestimation of the ConfTr objective function. Through extensive experiments on\nvarious benchmark datasets, we demonstrate that VR-ConfTr consistently achieves\nfaster convergence and smaller prediction sets compared to baselines.\n", "date": "2024-11-03", "categories": []} {"id": "2501.10139", "title": "Conformal Prediction Sets with Improved Conditional Coverage using Trust\n Scores", "abstract": " Standard conformal prediction offers a marginal guarantee on coverage, but\nfor prediction sets to be truly useful, they should ideally ensure coverage\nconditional on each test point. Unfortunately, it is impossible to achieve\nexact, distribution-free conditional coverage in finite samples. In this work,\nwe propose an alternative conformal prediction algorithm that targets coverage\nwhere it matters most--in instances where a classifier is overconfident in its\nincorrect predictions. We start by dissecting miscoverage events in\nmarginally-valid conformal prediction, and show that miscoverage rates vary\nbased on the classifier's confidence and its deviation from the Bayes optimal\nclassifier. Motivated by this insight, we develop a variant of conformal\nprediction that targets coverage conditional on a reduced set of two variables:\nthe classifier's confidence in a prediction and a nonparametric trust score\nthat measures its deviation from the Bayes classifier. Empirical evaluation on\nmultiple image datasets shows that our method generally improves conditional\ncoverage properties compared to standard conformal prediction, including\nclass-conditional coverage, coverage over arbitrary subgroups, and coverage\nover demographic groups.\n", "date": "2025-01-17", "categories": []} {"id": "2502.03609", "title": "Multivariate Conformal Prediction using Optimal Transport", "abstract": " Conformal prediction (CP) quantifies the uncertainty of machine learning\nmodels by constructing sets of plausible outputs. These sets are constructed by\nleveraging a so-called conformity score, a quantity computed using the input\npoint of interest, a prediction model, and past observations. CP sets are then\nobtained by evaluating the conformity score of all possible outputs, and\nselecting them according to the rank of their scores. Due to this ranking step,\nmost CP approaches rely on a score functions that are univariate. The challenge\nin extending these scores to multivariate spaces lies in the fact that no\ncanonical order for vectors exists. To address this, we leverage a natural\nextension of multivariate score ranking based on optimal transport (OT). Our\nmethod, OTCP, offers a principled framework for constructing conformal\nprediction sets in multidimensional settings, preserving distribution-free\ncoverage guarantees with finite data samples. We demonstrate tangible gains in\na benchmark dataset of multivariate regression problems and address\ncomputational \\& statistical trade-offs that arise when estimating conformity\nscores through OT maps.\n", "date": "2025-02-05", "categories": []} {"id": "2406.12100", "title": "CUQDS: Conformal Uncertainty Quantification under Distribution Shift for\n Trajectory Prediction", "abstract": " Trajectory prediction models that can infer both finite future trajectories\nand their associated uncertainties of the target vehicles in an online setting\n(e.g., real-world application scenarios) is crucial for ensuring the safe and\nrobust navigation and path planning of autonomous vehicle motion. However, the\nmajority of existing trajectory prediction models have neither considered\nreducing the uncertainty as one objective during the training stage nor\nprovided reliable uncertainty quantification during inference stage under\npotential distribution shift. Therefore, in this paper, we propose the\nConformal Uncertainty Quantification under Distribution Shift framework, CUQDS,\nto quantify the uncertainty of the predicted trajectories of existing\ntrajectory prediction models under potential data distribution shift, while\nconsidering improving the prediction accuracy of the models and reducing the\nestimated uncertainty during the training stage. Specifically, CUQDS includes\n1) a learning-based Gaussian process regression module that models the output\ndistribution of the base model (any existing trajectory prediction or time\nseries forecasting neural networks) and reduces the estimated uncertainty by\nadditional loss term, and 2) a statistical-based Conformal P control module to\ncalibrate the estimated uncertainty from the Gaussian process regression module\nin an online setting under potential distribution shift between training and\ntesting data.\n", "date": "2024-06-17", "categories": []} {"id": "2502.02561", "title": "Decision Theoretic Foundations for Conformal Prediction: Optimal\n Uncertainty Quantification for Risk-Averse Agents", "abstract": " A fundamental question in data-driven decision making is how to quantify the\nuncertainty of predictions in ways that can usefully inform downstream action.\nThis interface between prediction uncertainty and decision-making is especially\nimportant in risk-sensitive domains, such as medicine. In this paper, we\ndevelop decision-theoretic foundations that connect uncertainty quantification\nusing prediction sets with risk-averse decision-making. Specifically, we answer\nthree fundamental questions: (1) What is the correct notion of uncertainty\nquantification for risk-averse decision makers? We prove that prediction sets\nare optimal for decision makers who wish to optimize their value at risk. (2)\nWhat is the optimal policy that a risk averse decision maker should use to map\nprediction sets to actions? We show that a simple max-min decision policy is\noptimal for risk-averse decision makers. Finally, (3) How can we derive\nprediction sets that are optimal for such decision makers? We provide an exact\ncharacterization in the population regime and a distribution free finite-sample\nconstruction. Answering these questions naturally leads to an algorithm,\nRisk-Averse Calibration (RAC), which follows a provably optimal design for\nderiving action policies from predictions. RAC is designed to be both\npractical-capable of leveraging the quality of predictions in a black-box\nmanner to enhance downstream utility-and safe-adhering to a user-defined risk\nthreshold and optimizing the corresponding risk quantile of the user's\ndownstream utility. Finally, we experimentally demonstrate the significant\nadvantages of RAC in applications such as medical diagnosis and recommendation\nsystems. Specifically, we show that RAC achieves a substantially improved\ntrade-off between safety and utility, offering higher utility compared to\nexisting methods while maintaining the safety guarantee.\n", "date": "2025-02-04", "categories": []} {"id": "2501.10533", "title": "A Unified Comparative Study with Generalized Conformity Scores for\n Multi-Output Conformal Regression", "abstract": " Conformal prediction provides a powerful framework for constructing\ndistribution-free prediction regions with finite-sample coverage guarantees.\nWhile extensively studied in univariate settings, its extension to multi-output\nproblems presents additional challenges, including complex output dependencies\nand high computational costs, and remains relatively underexplored. In this\nwork, we present a unified comparative study of nine conformal methods with\ndifferent multivariate base models for constructing multivariate prediction\nregions within the same framework. This study highlights their key properties\nwhile also exploring the connections between them. Additionally, we introduce\ntwo novel classes of conformity scores for multi-output regression that\ngeneralize their univariate counterparts. These scores ensure asymptotic\nconditional coverage while maintaining exact finite-sample marginal coverage.\nOne class is compatible with any generative model, offering broad\napplicability, while the other is computationally efficient, leveraging the\nproperties of invertible generative models. Finally, we conduct a comprehensive\nempirical evaluation across 13 tabular datasets, comparing all the multi-output\nconformal methods explored in this work. To ensure a fair and consistent\ncomparison, all methods are implemented within a unified code base.\n", "date": "2025-01-17", "categories": []} {"id": "2312.06531", "title": "Uncertainty quantification in automated valuation models with spatially\n weighted conformal prediction", "abstract": " Non-parametric machine learning models, such as random forests and gradient\nboosted trees, are frequently used to estimate house prices due to their\npredictive accuracy, but a main drawback of such methods is their limited\nability to quantify prediction uncertainty. Conformal prediction (CP) is a\nmodel-agnostic framework for constructing confidence sets around predictions of\nmachine learning models with minimal assumptions. However, due to the spatial\ndependencies observed in house prices, direct application of CP leads to\nconfidence sets that are not calibrated everywhere, i.e., the confidence sets\nwill be too large in certain geographical regions and too small in others. We\nsurvey various approaches to adjust the CP confidence set to account for this\nand demonstrate their performance on a data set from the housing market in\nOslo, Norway. Our findings indicate that calibrating the confidence sets on a\nspatially weighted version of the non-conformity scores makes the coverage more\nconsistently calibrated across geographical regions. We also perform a\nsimulation study on synthetically generated sale prices to empirically explore\nthe performance of CP on housing market data under idealized conditions with\nknown data-generating mechanisms.\n", "date": "2023-12-11", "categories": []} {"id": "2501.18060", "title": "Noise-Adaptive Conformal Classification with Marginal Coverage", "abstract": " Conformal inference provides a rigorous statistical framework for uncertainty\nquantification in machine learning, enabling well-calibrated prediction sets\nwith precise coverage guarantees for any classification model. However, its\nreliance on the idealized assumption of perfect data exchangeability limits its\neffectiveness in the presence of real-world complications, such as low-quality\nlabels -- a widespread issue in modern large-scale data sets. This work tackles\nthis open problem by introducing an adaptive conformal inference method capable\nof efficiently handling deviations from exchangeability caused by random label\nnoise, leading to informative prediction sets with tight marginal coverage\nguarantees even in those challenging scenarios. We validate our method through\nextensive numerical experiments demonstrating its effectiveness on synthetic\nand real data sets, including CIFAR-10H and BigEarthNet.\n", "date": "2025-01-29", "categories": []} {"id": "2303.02770", "title": "Universal distribution of the empirical coverage in split conformal\n prediction", "abstract": " When split conformal prediction operates in batch mode with exchangeable\ndata, we determine the exact distribution of the empirical coverage of\nprediction sets produced for a finite batch of future observables, as well as\nthe exact distribution of its almost sure limit when the batch size goes to\ninfinity. Both distributions are universal, being determined solely by the\nnominal miscoverage level and the calibration sample size, thereby establishing\na criterion for choosing the minimum required calibration sample size in\napplications.\n", "date": "2023-03-05", "categories": []} {"id": "2501.11413", "title": "Generalization and Informativeness of Weighted Conformal Risk Control\n Under Covariate Shift", "abstract": " Predictive models are often required to produce reliable predictions under\nstatistical conditions that are not matched to the training data. A common type\nof training-testing mismatch is covariate shift, where the conditional\ndistribution of the target variable given the input features remains fixed,\nwhile the marginal distribution of the inputs changes. Weighted conformal risk\ncontrol (W-CRC) uses data collected during the training phase to convert point\npredictions into prediction sets with valid risk guarantees at test time\ndespite the presence of a covariate shift. However, while W-CRC provides\nstatistical reliability, its efficiency -- measured by the size of the\nprediction sets -- can only be assessed at test time. In this work, we relate\nthe generalization properties of the base predictor to the efficiency of W-CRC\nunder covariate shifts. Specifically, we derive a bound on the inefficiency of\nthe W-CRC predictor that depends on algorithmic hyperparameters and\ntask-specific quantities available at training time. This bound offers insights\non relationships between the informativeness of the prediction sets, the extent\nof the covariate shift, and the size of the calibration and training sets.\nExperiments on fingerprinting-based localization validate the theoretical\nresults.\n", "date": "2025-01-20", "categories": []} {"id": "2501.08963", "title": "Training-Aware Risk Control for Intensity Modulated Radiation Therapies\n Quality Assurance with Conformal Prediction", "abstract": " Measurement quality assurance (QA) practices play a key role in the safe use\nof Intensity Modulated Radiation Therapies (IMRT) for cancer treatment. These\npractices have reduced measurement-based IMRT QA failure below 1%. However,\nthese practices are time and labor intensive which can lead to delays in\npatient care. In this study, we examine how conformal prediction methodologies\ncan be used to robustly triage plans. We propose a new training-aware conformal\nrisk control method by combining the benefit of conformal risk control and\nconformal training. We incorporate the decision making thresholds based on the\ngamma passing rate, along with the risk functions used in clinical evaluation,\ninto the design of the risk control framework. Our method achieves high\nsensitivity and specificity and significantly reduces the number of plans\nneeding measurement without generating a huge confidence interval. Our results\ndemonstrate the validity and applicability of conformal prediction methods for\nimproving efficiency and reducing the workload of the IMRT QA process.\n", "date": "2025-01-15", "categories": []} {"id": "2501.06308", "title": "Uncertainty Estimation for Path Loss and Radio Metric Models", "abstract": " This research leverages Conformal Prediction (CP) in the form of Conformal\nPredictive Systems (CPS) to accurately estimate uncertainty in a suite of\nmachine learning (ML)-based radio metric models [1] as well as in a 2-D\nmap-based ML path loss model [2]. Utilizing diverse difficulty estimators, we\nconstruct 95% confidence prediction intervals (PIs) that are statistically\nrobust. Our experiments demonstrate that CPS models, trained on Toronto\ndatasets, generalize effectively to other cities such as Vancouver and\nMontreal, maintaining high coverage and reliability. Furthermore, the employed\ndifficulty estimators identify challenging samples, leading to measurable\nreductions in RMSE as dataset difficulty decreases. These findings highlight\nthe effectiveness of scalable and reliable uncertainty estimation through CPS\nin wireless network modeling, offering important potential insights for network\nplanning, operations, and spectrum management.\n", "date": "2025-01-10", "categories": []} {"id": "2501.07185", "title": "Uncertainty Guarantees on Automated Precision Weeding using Conformal\n Prediction", "abstract": " Precision agriculture in general, and precision weeding in particular, have\ngreatly benefited from the major advancements in deep learning and computer\nvision. A large variety of commercial robotic solutions are already available\nand deployed. However, the adoption by farmers of such solutions is still low\nfor many reasons, an important one being the lack of trust in these systems.\nThis is in great part due to the opaqueness and complexity of deep neural\nnetworks and the manufacturers' inability to provide valid guarantees on their\nperformance. Conformal prediction, a well-established methodology in the\nmachine learning community, is an efficient and reliable strategy for providing\ntrustworthy guarantees on the predictions of any black-box model under very\nminimal constraints. Bridging the gap between the safe machine learning and\nprecision agriculture communities, this article showcases conformal prediction\nin action on the task of precision weeding through deep learning-based image\nclassification. After a detailed presentation of the conformal prediction\nmethodology and the development of a precision spraying pipeline based on a\n''conformalized'' neural network and well-defined spraying decision rules, the\narticle evaluates this pipeline on two real-world scenarios: one under\nin-distribution conditions, the other reflecting a near out-of-distribution\nsetting. The results show that we are able to provide formal, i.e. certifiable,\nguarantees on spraying at least 90% of the weeds.\n", "date": "2025-01-13", "categories": []} {"id": "2406.05405", "title": "Robust Conformal Prediction Using Privileged Information", "abstract": " We develop a method to generate prediction sets with a guaranteed coverage\nrate that is robust to corruptions in the training data, such as missing or\nnoisy variables. Our approach builds on conformal prediction, a powerful\nframework to construct prediction sets that are valid under the i.i.d\nassumption. Importantly, naively applying conformal prediction does not provide\nreliable predictions in this setting, due to the distribution shift induced by\nthe corruptions. To account for the distribution shift, we assume access to\nprivileged information (PI). The PI is formulated as additional features that\nexplain the distribution shift, however, they are only available during\ntraining and absent at test time. We approach this problem by introducing a\nnovel generalization of weighted conformal prediction and support our method\nwith theoretical coverage guarantees. Empirical experiments on both real and\nsynthetic datasets indicate that our approach achieves a valid coverage rate\nand constructs more informative predictions compared to existing methods, which\nare not supported by theoretical guarantees.\n", "date": "2024-06-08", "categories": []} {"id": "2501.02767", "title": "Enhancing Trustworthiness of Graph Neural Networks with Rank-Based\n Conformal Training", "abstract": " Graph Neural Networks (GNNs) has been widely used in a variety of fields\nbecause of their great potential in representing graph-structured data.\nHowever, lacking of rigorous uncertainty estimations limits their application\nin high-stakes. Conformal Prediction (CP) can produce statistically guaranteed\nuncertainty estimates by using the classifier's probability estimates to obtain\nprediction sets, which contains the true class with a user-specified\nprobability. In this paper, we propose a Rank-based CP during training\nframework to GNNs (RCP-GNN) for reliable uncertainty estimates to enhance the\ntrustworthiness of GNNs in the node classification scenario. By exploiting rank\ninformation of the classifier's outcome, prediction sets with desired coverage\nrate can be efficiently constructed. The strategy of CP during training with\ndifferentiable rank-based conformity loss function is further explored to adapt\nprediction sets according to network topology information. In this way, the\ncomposition of prediction sets can be guided by the goal of jointly reducing\ninefficiency and probability estimation errors. Extensive experiments on\nseveral real-world datasets show that our model achieves any pre-defined target\nmarginal coverage while significantly reducing the inefficiency compared with\nstate-of-the-art methods.\n", "date": "2025-01-06", "categories": []} {"id": "2407.14495", "title": "Conformal Thresholded Intervals for Efficient Regression", "abstract": " This paper introduces Conformal Thresholded Intervals (CTI), a novel\nconformal regression method that aims to produce the smallest possible\nprediction set with guaranteed coverage. Unlike existing methods that rely on\nnested conformal frameworks and full conditional distribution estimation, CTI\nestimates the conditional probability density for a new response to fall into\neach interquantile interval using off-the-shelf multi-output quantile\nregression. By leveraging the inverse relationship between interval length and\nprobability density, CTI constructs prediction sets by thresholding the\nestimated conditional interquantile intervals based on their length. The\noptimal threshold is determined using a calibration set to ensure marginal\ncoverage, effectively balancing the trade-off between prediction set size and\ncoverage. CTI's approach is computationally efficient and avoids the complexity\nof estimating the full conditional distribution. The method is theoretically\ngrounded, with provable guarantees for marginal coverage and achieving the\nsmallest prediction size given by Neyman-Pearson . Extensive experimental\nresults demonstrate that CTI achieves superior performance compared to\nstate-of-the-art conformal regression methods across various datasets,\nconsistently producing smaller prediction sets while maintaining the desired\ncoverage level. The proposed method offers a simple yet effective solution for\nreliable uncertainty quantification in regression tasks, making it an\nattractive choice for practitioners seeking accurate and efficient conformal\nprediction.\n", "date": "2024-07-19", "categories": []} {"id": "2408.10939", "title": "Conformalized Interval Arithmetic with Symmetric Calibration", "abstract": " Uncertainty quantification is essential in decision-making, especially when\njoint distributions of random variables are involved. While conformal\nprediction provides distribution-free prediction sets with valid coverage\nguarantees, it traditionally focuses on single predictions. This paper\nintroduces novel conformal prediction methods for estimating the sum or average\nof unknown labels over specific index sets. We develop conformal prediction\nintervals for single target to the prediction interval for sum of multiple\ntargets. Under permutation invariant assumptions, we prove the validity of our\nproposed method. We also apply our algorithms on class average estimation and\npath cost prediction tasks, and we show that our method outperforms existing\nconformalized approaches as well as non-conformal approaches.\n", "date": "2024-08-20", "categories": []} {"id": "2412.19318", "title": "Adaptive Conformal Inference by Betting", "abstract": " Conformal prediction is a valuable tool for quantifying predictive\nuncertainty of machine learning models. However, its applicability relies on\nthe assumption of data exchangeability, a condition which is often not met in\nreal-world scenarios. In this paper, we consider the problem of adaptive\nconformal inference without any assumptions about the data generating process.\nExisting approaches for adaptive conformal inference are based on optimizing\nthe pinball loss using variants of online gradient descent. A notable\nshortcoming of such approaches is in their explicit dependence on and\nsensitivity to the choice of the learning rates. In this paper, we propose a\ndifferent approach for adaptive conformal inference that leverages\nparameter-free online convex optimization techniques. We prove that our method\ncontrols long-term miscoverage frequency at a nominal level and demonstrate its\nconvincing empirical performance without any need of performing cumbersome\nparameter tuning.\n", "date": "2024-12-26", "categories": []} {"id": "2412.18144", "title": "Neural Conformal Control for Time Series Forecasting", "abstract": " We introduce a neural network conformal prediction method for time series\nthat enhances adaptivity in non-stationary environments. Our approach acts as a\nneural controller designed to achieve desired target coverage, leveraging\nauxiliary multi-view data with neural network encoders in an end-to-end manner\nto further enhance adaptivity. Additionally, our model is designed to enhance\nthe consistency of prediction intervals in different quantiles by integrating\nmonotonicity constraints and leverages data from related tasks to boost\nfew-shot learning performance. Using real-world datasets from epidemics,\nelectric demand, weather, and others, we empirically demonstrate significant\nimprovements in coverage and probabilistic accuracy, and find that our method\nis the only one that combines good calibration with consistency in prediction\nintervals.\n", "date": "2024-12-24", "categories": []} {"id": "2402.04344", "title": "Does confidence calibration improve conformal prediction?", "abstract": " Conformal prediction is an emerging technique for uncertainty quantification\nthat constructs prediction sets guaranteed to contain the true label with a\npredefined probability. Previous works often employ temperature scaling to\ncalibrate classifiers, assuming that confidence calibration benefits conformal\nprediction. However, the specific impact of confidence calibration on conformal\nprediction remains underexplored. In this work, we make two key discoveries\nabout the impact of confidence calibration methods on adaptive conformal\nprediction. Firstly, we empirically show that current confidence calibration\nmethods (e.g., temperature scaling) typically lead to larger prediction sets in\nadaptive conformal prediction. Secondly, by investigating the role of\ntemperature value, we observe that high-confidence predictions can enhance the\nefficiency of adaptive conformal prediction. Theoretically, we prove that\npredictions with higher confidence result in smaller prediction sets on\nexpectation. This finding implies that the rescaling parameters in these\ncalibration methods, when optimized with cross-entropy loss, might counteract\nthe goal of generating efficient prediction sets. To address this issue, we\npropose Conformal Temperature Scaling (ConfTS), a variant of temperature\nscaling with a novel loss function designed to enhance the efficiency of\nprediction sets. This approach can be extended to optimize the parameters of\nother post-hoc methods of confidence calibration. Extensive experiments\ndemonstrate that our method improves existing adaptive conformal prediction\nmethods in classification tasks, especially with LLMs.\n", "date": "2024-02-06", "categories": []} {"id": "2412.10459", "title": "Conformal Prediction on Quantifying Uncertainty of Dynamic Systems", "abstract": " Numerous studies have focused on learning and understanding the dynamics of\nphysical systems from video data, such as spatial intelligence. Artificial\nintelligence requires quantitative assessments of the uncertainty of the model\nto ensure reliability. However, there is still a relative lack of systematic\nassessment of the uncertainties, particularly the uncertainties of the physical\ndata. Our motivation is to introduce conformal prediction into the uncertainty\nassessment of dynamical systems, providing a method supported by theoretical\nguarantees. This paper uses the conformal prediction method to assess\nuncertainties with benchmark operator learning methods. We have also compared\nthe Monte Carlo Dropout and Ensemble methods in the partial differential\nequations dataset, effectively evaluating uncertainty through straight\nroll-outs, making it ideal for time-series tasks.\n", "date": "2024-12-12", "categories": []} {"id": "2412.13159", "title": "A Conformal Approach to Feature-based Newsvendor under Model\n Misspecification", "abstract": " In many data-driven decision-making problems, performance guarantees often\ndepend heavily on the correctness of model assumptions, which may frequently\nfail in practice. We address this issue in the context of a feature-based\nnewsvendor problem, where demand is influenced by observed features such as\ndemographics and seasonality. To mitigate the impact of model misspecification,\nwe propose a model-free and distribution-free framework inspired by conformal\nprediction. Our approach consists of two phases: a training phase, which can\nutilize any type of prediction method, and a calibration phase that\nconformalizes the model bias. To enhance predictive performance, we explore the\nbalance between data quality and quantity, recognizing the inherent trade-off:\nmore selective training data improves quality but reduces quantity.\nImportantly, we provide statistical guarantees for the conformalized critical\nquantile, independent of the correctness of the underlying model. Moreover, we\nquantify the confidence interval of the critical quantile, with its width\ndecreasing as data quality and quantity improve. We validate our framework\nusing both simulated data and a real-world dataset from the Capital Bikeshare\nprogram in Washington, D.C. Across these experiments, our proposed method\nconsistently outperforms benchmark algorithms, reducing newsvendor loss by up\nto 40% on the simulated data and 25% on the real-world dataset.\n", "date": "2024-12-17", "categories": []} {"id": "2406.18814", "title": "Length Optimization in Conformal Prediction", "abstract": " Conditional validity and length efficiency are two crucial aspects of\nconformal prediction (CP). Conditional validity ensures accurate uncertainty\nquantification for data subpopulations, while proper length efficiency ensures\nthat the prediction sets remain informative. Despite significant efforts to\naddress each of these issues individually, a principled framework that\nreconciles these two objectives has been missing in the CP literature. In this\npaper, we develop Conformal Prediction with Length-Optimization (CPL) - a novel\nand practical framework that constructs prediction sets with (near-) optimal\nlength while ensuring conditional validity under various classes of covariate\nshifts, including the key cases of marginal and group-conditional coverage. In\nthe infinite sample regime, we provide strong duality results which indicate\nthat CPL achieves conditional validity and length optimality. In the finite\nsample regime, we show that CPL constructs conditionally valid prediction sets.\nOur extensive empirical evaluations demonstrate the superior prediction set\nsize performance of CPL compared to state-of-the-art methods across diverse\nreal-world and synthetic datasets in classification, regression, and large\nlanguage model-based multiple choice question answering. An Implementation of\nour algorithm can be accessed at the following link:\nhttps://github.com/shayankiyani98/CP.\n", "date": "2024-06-27", "categories": []} {"id": "2406.06818", "title": "Conformal Prediction for Class-wise Coverage via Augmented Label Rank\n Calibration", "abstract": " Conformal prediction (CP) is an emerging uncertainty quantification framework\nthat allows us to construct a prediction set to cover the true label with a\npre-specified marginal or conditional probability. Although the valid coverage\nguarantee has been extensively studied for classification problems, CP often\nproduces large prediction sets which may not be practically useful. This issue\nis exacerbated for the setting of class-conditional coverage on imbalanced\nclassification tasks with many and/or imbalanced classes. This paper proposes\nthe Rank Calibrated Class-conditional CP (RC3P) algorithm to reduce the\nprediction set sizes to achieve class-conditional coverage, where the valid\ncoverage holds for each class. In contrast to the standard class-conditional CP\n(CCP) method that uniformly thresholds the class-wise conformity score for each\nclass, the augmented label rank calibration step allows RC3P to selectively\niterate this class-wise thresholding subroutine only for a subset of classes\nwhose class-wise top-k error is small. We prove that agnostic to the classifier\nand data distribution, RC3P achieves class-wise coverage. We also show that\nRC3P reduces the size of prediction sets compared to the CCP method.\nComprehensive experiments on multiple real-world datasets demonstrate that RC3P\nachieves class-wise coverage and 26.25% reduction in prediction set sizes on\naverage.\n", "date": "2024-06-10", "categories": []} {"id": "2412.04166", "title": "An In-Depth Examination of Risk Assessment in Multi-Class Classification\n Algorithms", "abstract": " Advanced classification algorithms are being increasingly used in\nsafety-critical applications like health-care, engineering, etc. In such\napplications, miss-classifications made by ML algorithms can result in\nsubstantial financial or health-related losses. To better anticipate and\nprepare for such losses, the algorithm user seeks an estimate for the\nprobability that the algorithm miss-classifies a sample. We refer to this task\nas the risk-assessment. For a variety of models and datasets, we numerically\nanalyze the performance of different methods in solving the risk-assessment\nproblem. We consider two solution strategies: a) calibration techniques that\ncalibrate the output probabilities of classification models to provide accurate\nprobability outputs; and b) a novel approach based upon the prediction interval\ngeneration technique of conformal prediction. Our conformal prediction based\napproach is model and data-distribution agnostic, simple to implement, and\nprovides reasonable results for a variety of use-cases. We compare the\ndifferent methods on a broad variety of models and datasets.\n", "date": "2024-12-05", "categories": []} {"id": "2412.03464", "title": "Validity and efficiency of the conformal CUSUM procedure", "abstract": " In this paper we study the validity and efficiency of a conformal version of\nthe CUSUM procedure for change detection both experimentally and theoretically.\n", "date": "2024-12-04", "categories": []} {"id": "2412.00653", "title": "Predictive Inference With Fast Feature Conformal Prediction", "abstract": " Conformal prediction is widely adopted in uncertainty quantification, due to\nits post-hoc, distribution-free, and model-agnostic properties. In the realm of\nmodern deep learning, researchers have proposed Feature Conformal Prediction\n(FCP), which deploys conformal prediction in a feature space, yielding reduced\nband lengths. However, the practical utility of FCP is limited due to the\ntime-consuming non-linear operations required to transform confidence bands\nfrom feature space to output space. In this paper, we introduce Fast Feature\nConformal Prediction (FFCP), which features a novel non-conformity score and is\nconvenient for practical applications. FFCP serves as a fast version of FCP, in\nthat it equivalently employs a Taylor expansion to approximate the\naforementioned non-linear operations in FCP. Empirical validations showcase\nthat FFCP performs comparably with FCP (both outperforming the vanilla version)\nwhile achieving a significant reduction in computational time by approximately\n50x. The code is available at https://github.com/ElvisWang1111/FastFeatureCP\n", "date": "2024-12-01", "categories": []} {"id": "2411.17983", "title": "Optimized Conformal Selection: Powerful Selective Inference After\n Conformity Score Optimization", "abstract": " Model selection/optimization in conformal inference is challenging, since it\nmay break the exchangeability between labeled and unlabeled data. We study this\nproblem in the context of conformal selection, which uses conformal p-values to\nselect ``interesting'' instances with large unobserved labels from a pool of\nunlabeled data, while controlling the FDR in finite sample. For validity,\nexisting solutions require the model choice to be independent of the data used\nto construct the p-values and calibrate the selection set. However, when\npresented with many model choices and limited labeled data, it is desirable to\n(i) select the best model in a data-driven manner, and (ii) mitigate power loss\ndue to sample splitting.\n This paper presents OptCS, a general framework that allows valid statistical\ntesting (selection) after flexible data-driven model optimization. We introduce\ngeneral conditions under which OptCS constructs valid conformal p-values\ndespite substantial data reuse and handles complex p-value dependencies to\nmaintain finite-sample FDR control via a novel multiple testing procedure. We\ninstantiate this general recipe to propose three FDR-controlling procedures,\neach optimizing the models differently: (i) selecting the most powerful one\namong multiple pre-trained candidate models, (ii) using all data for model\nfitting without sample splitting, and (iii) combining full-sample model fitting\nand selection. We demonstrate the efficacy of our methods via simulation\nstudies and real applications in drug discovery and alignment of large language\nmodels in radiology report generation.\n", "date": "2024-11-27", "categories": []} {"id": "2209.14295", "title": "Label Noise Robustness of Conformal Prediction", "abstract": " We study the robustness of conformal prediction, a powerful tool for\nuncertainty quantification, to label noise. Our analysis tackles both\nregression and classification problems, characterizing when and how it is\npossible to construct uncertainty sets that correctly cover the unobserved\nnoiseless ground truth labels. We further extend our theory and formulate the\nrequirements for correctly controlling a general loss function, such as the\nfalse negative proportion, with noisy labels. Our theory and experiments\nsuggest that conformal prediction and risk-controlling techniques with noisy\nlabels attain conservative risk over the clean ground truth labels whenever the\nnoise is dispersive and increases variability. In other adversarial cases, we\ncan also correct for noise of bounded size in the conformal prediction\nalgorithm in order to ensure achieving the correct risk of the ground truth\nlabels without score or data regularity.\n", "date": "2022-09-28", "categories": []} {"id": "2411.17042", "title": "Conformalised Conditional Normalising Flows for Joint Prediction Regions\n in time series", "abstract": " Conformal Prediction offers a powerful framework for quantifying uncertainty\nin machine learning models, enabling the construction of prediction sets with\nfinite-sample validity guarantees. While easily adaptable to non-probabilistic\nmodels, applying conformal prediction to probabilistic generative models, such\nas Normalising Flows is not straightforward. This work proposes a novel method\nto conformalise conditional normalising flows, specifically addressing the\nproblem of obtaining prediction regions for multi-step time series forecasting.\nOur approach leverages the flexibility of normalising flows to generate\npotentially disjoint prediction regions, leading to improved predictive\nefficiency in the presence of potential multimodal predictive distributions.\n", "date": "2024-11-26", "categories": []} {"id": "2411.17099", "title": "Spatio-Temporal Conformal Prediction for Power Outage Data", "abstract": " In recent years, increasingly unpredictable and severe global weather\npatterns have frequently caused long-lasting power outages. Building\nresilience, the ability to withstand, adapt to, and recover from major\ndisruptions, has become crucial for the power industry. To enable rapid\nrecovery, accurately predicting future outage numbers is essential. Rather than\nrelying on simple point estimates, we analyze extensive quarter-hourly outage\ndata and develop a graph conformal prediction method that delivers accurate\nprediction regions for outage numbers across the states for a time period. We\ndemonstrate the effectiveness of this method through extensive numerical\nexperiments in several states affected by extreme weather events that led to\nwidespread outages.\n", "date": "2024-11-26", "categories": []} {"id": "2405.18942", "title": "Verifiably Robust Conformal Prediction", "abstract": " Conformal Prediction (CP) is a popular uncertainty quantification method that\nprovides distribution-free, statistically valid prediction sets, assuming that\ntraining and test data are exchangeable. In such a case, CP's prediction sets\nare guaranteed to cover the (unknown) true test output with a user-specified\nprobability. Nevertheless, this guarantee is violated when the data is\nsubjected to adversarial attacks, which often result in a significant loss of\ncoverage. Recently, several approaches have been put forward to recover CP\nguarantees in this setting. These approaches leverage variations of randomised\nsmoothing to produce conservative sets which account for the effect of the\nadversarial perturbations. They are, however, limited in that they only support\n$\\ell^2$-bounded perturbations and classification tasks. This paper introduces\nVRCP (Verifiably Robust Conformal Prediction), a new framework that leverages\nrecent neural network verification methods to recover coverage guarantees under\nadversarial attacks. Our VRCP method is the first to support perturbations\nbounded by arbitrary norms including $\\ell^1$, $\\ell^2$, and $\\ell^\\infty$, as\nwell as regression tasks. We evaluate and compare our approach on image\nclassification tasks (CIFAR10, CIFAR100, and TinyImageNet) and regression tasks\nfor deep reinforcement learning environments. In every case, VRCP achieves\nabove nominal coverage and yields significantly more efficient and informative\nprediction regions than the SotA.\n", "date": "2024-05-29", "categories": []} {"id": "2403.16336", "title": "Predictive Inference in Multi-environment Scenarios", "abstract": " We address the challenge of constructing valid confidence intervals and sets\nin problems of prediction across multiple environments. We investigate two\ntypes of coverage suitable for these problems, extending the jackknife and\nsplit-conformal methods to show how to obtain distribution-free coverage in\nsuch non-traditional, potentially hierarchical data-generating scenarios. We\ndemonstrate a novel resizing method to adapt to problem difficulty, which\napplies both to existing approaches for predictive inference and the methods we\ndevelop; this reduces prediction set sizes using limited information from the\ntest environment, a key to the methods' practical performance, which we\nevaluate through neurochemical sensing and species classification datasets. Our\ncontributions also include extensions for settings with non-real-valued\nresponses, a theory of consistency for predictive inference in these general\nproblems, and insights on the limits of conditional coverage.\n", "date": "2024-03-25", "categories": []} {"id": "2411.03678", "title": "Multi-model Ensemble Conformal Prediction in Dynamic Environments", "abstract": " Conformal prediction is an uncertainty quantification method that constructs\na prediction set for a previously unseen datum, ensuring the true label is\nincluded with a predetermined coverage probability. Adaptive conformal\nprediction has been developed to address data distribution shifts in dynamic\nenvironments. However, the efficiency of prediction sets varies depending on\nthe learning model used. Employing a single fixed model may not consistently\noffer the best performance in dynamic environments with unknown data\ndistribution shifts. To address this issue, we introduce a novel adaptive\nconformal prediction framework, where the model used for creating prediction\nsets is selected on the fly from multiple candidate models. The proposed\nalgorithm is proven to achieve strongly adaptive regret over all intervals\nwhile maintaining valid coverage. Experiments on real and synthetic datasets\ncorroborate that the proposed approach consistently yields more efficient\nprediction sets while maintaining valid coverage, outperforming alternative\nmethods.\n", "date": "2024-11-06", "categories": []} {"id": "2405.10301", "title": "Conformal Alignment: Knowing When to Trust Foundation Models with\n Guarantees", "abstract": " Before deploying outputs from foundation models in high-stakes tasks, it is\nimperative to ensure that they align with human values. For instance, in\nradiology report generation, reports generated by a vision-language model must\nalign with human evaluations before their use in medical decision-making. This\npaper presents Conformal Alignment, a general framework for identifying units\nwhose outputs meet a user-specified alignment criterion. It is guaranteed that\non average, a prescribed fraction of selected units indeed meet the alignment\ncriterion, regardless of the foundation model or the data distribution. Given\nany pre-trained model and new units with model-generated outputs, Conformal\nAlignment leverages a set of reference data with ground-truth alignment status\nto train an alignment predictor. It then selects new units whose predicted\nalignment scores surpass a data-dependent threshold, certifying their\ncorresponding outputs as trustworthy. Through applications to question\nanswering and radiology report generation, we demonstrate that our method is\nable to accurately identify units with trustworthy outputs via lightweight\ntraining over a moderate amount of reference data. En route, we investigate the\ninformativeness of various features in alignment prediction and combine them\nwith standard models to construct the alignment predictor.\n", "date": "2024-05-16", "categories": []} {"id": "2406.18777", "title": "Aligning Model Properties via Conformal Risk Control", "abstract": " AI model alignment is crucial due to inadvertent biases in training data and\nthe underspecified machine learning pipeline, where models with excellent test\nmetrics may not meet end-user requirements. While post-training alignment via\nhuman feedback shows promise, these methods are often limited to generative AI\nsettings where humans can interpret and provide feedback on model outputs. In\ntraditional non-generative settings with numerical or categorical outputs,\ndetecting misalignment through single-sample outputs remains challenging, and\nenforcing alignment during training requires repeating costly training\nprocesses. In this paper we consider an alternative strategy. We propose\ninterpreting model alignment through property testing, defining an aligned\nmodel $f$ as one belonging to a subset $\\mathcal{P}$ of functions that exhibit\nspecific desired behaviors. We focus on post-processing a pre-trained model $f$\nto better align with $\\mathcal{P}$ using conformal risk control. Specifically,\nwe develop a general procedure for converting queries for testing a given\nproperty $\\mathcal{P}$ to a collection of loss functions suitable for use in a\nconformal risk control algorithm. We prove a probabilistic guarantee that the\nresulting conformal interval around $f$ contains a function approximately\nsatisfying $\\mathcal{P}$. We exhibit applications of our methodology on a\ncollection of supervised learning datasets for (shape-constrained) properties\nsuch as monotonicity and concavity. The general procedure is flexible and can\nbe applied to a wide range of desired properties. Finally, we prove that\npre-trained models will always require alignment techniques even as model sizes\nor training data increase, as long as the training data contains even small\nbiases.\n", "date": "2024-06-26", "categories": []} {"id": "2411.01558", "title": "Adaptive Conformal Inference by Particle Filtering under Hidden Markov\n Models", "abstract": " Conformal inference is a statistical method used to construct prediction sets\nfor point predictors, providing reliable uncertainty quantification with\nprobability guarantees. This method utilizes historical labeled data to\nestimate the conformity or nonconformity between predictions and true labels.\nHowever, conducting conformal inference for hidden states under hidden Markov\nmodels (HMMs) presents a significant challenge, as the hidden state data is\nunavailable, resulting in the absence of a true label set to serve as a\nconformal calibration set. This paper proposes an adaptive conformal inference\nframework that leverages a particle filtering approach to address this issue.\nRather than directly focusing on the unobservable hidden state, we innovatively\nuse weighted particles as an approximation of the actual posterior distribution\nof the hidden state. Our goal is to produce prediction sets that encompass\nthese particles to achieve a specific aggregate weight sum, referred to as the\naggregated coverage level. The proposed framework can adapt online to the\ntime-varying distribution of data and achieve the defined marginal aggregated\ncoverage level in both one-step and multi-step inference over the long term. We\nverify the effectiveness of this approach through a real-time target\nlocalization simulation study.\n", "date": "2024-11-03", "categories": []} {"id": "2411.01596", "title": "Strategic Conformal Prediction", "abstract": " When a machine learning model is deployed, its predictions can alter its\nenvironment, as better informed agents strategize to suit their own interests.\nWith such alterations in mind, existing approaches to uncertainty\nquantification break. In this work we propose a new framework, Strategic\nConformal Prediction, which is capable of robust uncertainty quantification in\nsuch a setting. Strategic Conformal Prediction is backed by a series of\ntheoretical guarantees spanning marginal coverage, training-conditional\ncoverage, tightness and robustness to misspecification that hold in a\ndistribution-free manner. Experimental analysis further validates our method,\nshowing its remarkable effectiveness in face of arbitrary strategic\nalterations, whereas other methods break.\n", "date": "2024-11-03", "categories": []} {"id": "2402.07307", "title": "Self-Calibrating Conformal Prediction", "abstract": " In machine learning, model calibration and predictive inference are essential\nfor producing reliable predictions and quantifying uncertainty to support\ndecision-making. Recognizing the complementary roles of point and interval\npredictions, we introduce Self-Calibrating Conformal Prediction, a method that\ncombines Venn-Abers calibration and conformal prediction to deliver calibrated\npoint predictions alongside prediction intervals with finite-sample validity\nconditional on these predictions. To achieve this, we extend the original\nVenn-Abers procedure from binary classification to regression. Our theoretical\nframework supports analyzing conformal prediction methods that involve\ncalibrating model predictions and subsequently constructing conditionally valid\nprediction intervals on the same data, where the conditioning set or conformity\nscores may depend on the calibrated predictions. Real-data experiments show\nthat our method improves interval efficiency through model calibration and\noffers a practical alternative to feature-conditional validity.\n", "date": "2024-02-11", "categories": []} {"id": "2406.09714", "title": "Large language model validity via enhanced conformal prediction methods", "abstract": " We develop new conformal inference methods for obtaining validity guarantees\non the output of large language models (LLMs). Prior work in conformal language\nmodeling identifies a subset of the text that satisfies a high-probability\nguarantee of correctness. These methods work by filtering claims from the LLM's\noriginal response if a scoring function evaluated on the claim fails to exceed\na threshold calibrated via split conformal prediction. Existing methods in this\narea suffer from two deficiencies. First, the guarantee stated is not\nconditionally valid. The trustworthiness of the filtering step may vary based\non the topic of the response. Second, because the scoring function is\nimperfect, the filtering step can remove many valuable and accurate claims. We\naddress both of these challenges via two new conformal methods. First, we\ngeneralize the conditional conformal procedure of Gibbs et al. (2023) in order\nto adaptively issue weaker guarantees when they are required to preserve the\nutility of the output. Second, we show how to systematically improve the\nquality of the scoring function via a novel algorithm for differentiating\nthrough the conditional conformal procedure. We demonstrate the efficacy of our\napproach on biography and medical question-answering datasets.\n", "date": "2024-06-14", "categories": []} {"id": "2406.10490", "title": "Active, anytime-valid risk controlling prediction sets", "abstract": " Rigorously establishing the safety of black-box machine learning models\nconcerning critical risk measures is important for providing guarantees about\nmodel behavior. Recently, Bates et. al. (JACM '24) introduced the notion of a\nrisk controlling prediction set (RCPS) for producing prediction sets that are\nstatistically guaranteed low risk from machine learning models. Our method\nextends this notion to the sequential setting, where we provide guarantees even\nwhen the data is collected adaptively, and ensures that the risk guarantee is\nanytime-valid, i.e., simultaneously holds at all time steps. Further, we\npropose a framework for constructing RCPSes for active labeling, i.e., allowing\none to use a labeling policy that chooses whether to query the true label for\neach received data point and ensures that the expected proportion of data\npoints whose labels are queried are below a predetermined label budget. We also\ndescribe how to use predictors (i.e., the machine learning model for which we\nprovide risk control guarantees) to further improve the utility of our RCPSes\nby estimating the expected risk conditioned on the covariates. We characterize\nthe optimal choices of label policy and predictor under a fixed label budget\nand show a regret result that relates the estimation error of the optimal\nlabeling policy and predictor to the wealth process that underlies our RCPSes.\nLastly, we present practical ways of formulating label policies and empirically\nshow that our label policies use fewer labels to reach higher utility than\nnaive baseline labeling strategies on both simulations and real data.\n", "date": "2024-06-15", "categories": []} {"id": "2405.15106", "title": "Conformal Classification with Equalized Coverage for Adaptively Selected\n Groups", "abstract": " This paper introduces a conformal inference method to evaluate uncertainty in\nclassification by generating prediction sets with valid coverage conditional on\nadaptively chosen features. These features are carefully selected to reflect\npotential model limitations or biases. This can be useful to find a practical\ncompromise between efficiency -- by providing informative predictions -- and\nalgorithmic fairness -- by ensuring equalized coverage for the most sensitive\ngroups. We demonstrate the validity and effectiveness of this method on\nsimulated and real data sets.\n", "date": "2024-05-23", "categories": []} {"id": "2402.16300", "title": "Conformalized Selective Regression", "abstract": " Should prediction models always deliver a prediction? In the pursuit of\nmaximum predictive performance, critical considerations of reliability and\nfairness are often overshadowed, particularly when it comes to the role of\nuncertainty. Selective regression, also known as the \"reject option,\" allows\nmodels to abstain from predictions in cases of considerable uncertainty.\nInitially proposed seven decades ago, approaches to selective regression have\nmostly focused on distribution-based proxies for measuring uncertainty,\nparticularly conditional variance. However, this focus neglects the significant\ninfluence of model-specific biases on a model's performance. In this paper, we\npropose a novel approach to selective regression by leveraging conformal\nprediction, which provides grounded confidence measures for individual\npredictions based on model-specific biases. In addition, we propose a\nstandardized evaluation framework to allow proper comparison of selective\nregression approaches. Via an extensive experimental approach, we demonstrate\nhow our proposed approach, conformalized selective regression, demonstrates an\nadvantage over multiple state-of-the-art baselines.\n", "date": "2024-02-26", "categories": []} {"id": "2402.14080", "title": "Efficient Normalized Conformal Prediction and Uncertainty Quantification\n for Anti-Cancer Drug Sensitivity Prediction with Deep Regression Forests", "abstract": " Deep learning models are being adopted and applied on various critical\ndecision-making tasks, yet they are trained to provide point predictions\nwithout providing degrees of confidence. The trustworthiness of deep learning\nmodels can be increased if paired with uncertainty estimations. Conformal\nPrediction has emerged as a promising method to pair machine learning models\nwith prediction intervals, allowing for a view of the model's uncertainty.\nHowever, popular uncertainty estimation methods for conformal prediction fail\nto provide heteroskedastic intervals that are equally accurate for all samples.\nIn this paper, we propose a method to estimate the uncertainty of each sample\nby calculating the variance obtained from a Deep Regression Forest. We show\nthat the deep regression forest variance improves the efficiency and coverage\nof normalized inductive conformal prediction on a drug response prediction\ntask.\n", "date": "2024-02-21", "categories": []} {"id": "2410.19077", "title": "Target Strangeness: A Novel Conformal Prediction Difficulty Estimator", "abstract": " This paper introduces Target Strangeness, a novel difficulty estimator for\nconformal prediction (CP) that offers an alternative approach for normalizing\nprediction intervals (PIs). By assessing how atypical a prediction is within\nthe context of its nearest neighbours' target distribution, Target Strangeness\ncan surpass the current state-of-the-art performance. This novel difficulty\nestimator is evaluated against others in the context of several conformal\nregression experiments.\n", "date": "2024-10-24", "categories": []} {"id": "2404.15018", "title": "Conformal Predictive Systems Under Covariate Shift", "abstract": " Conformal Predictive Systems (CPS) offer a versatile framework for\nconstructing predictive distributions, allowing for calibrated inference and\ninformative decision-making. However, their applicability has been limited to\nscenarios adhering to the Independent and Identically Distributed (IID) model\nassumption. This paper extends CPS to accommodate scenarios characterized by\ncovariate shifts. We therefore propose Weighted CPS (WCPS), akin to Weighted\nConformal Prediction (WCP), leveraging likelihood ratios between training and\ntesting covariate distributions. This extension enables the construction of\nnonparametric predictive distributions capable of handling covariate shifts. We\npresent theoretical underpinnings and conjectures regarding the validity and\nefficacy of WCPS and demonstrate its utility through empirical evaluations on\nboth synthetic and real-world datasets. Our simulation experiments indicate\nthat WCPS are probabilistically calibrated under covariate shift.\n", "date": "2024-04-23", "categories": []} {"id": "2410.12297", "title": "Conjunction Subspaces Test for Conformal and Selective Classification", "abstract": " In this paper, we present a new classifier, which integrates significance\ntesting results over different random subspaces to yield consensus p-values for\nquantifying the uncertainty of classification decision. The null hypothesis is\nthat the test sample has no association with the target class on a randomly\nchosen subspace, and hence the classification problem can be formulated as a\nproblem of testing for the conjunction of hypotheses. The proposed classifier\ncan be easily deployed for the purpose of conformal prediction and selective\nclassification with reject and refine options by simply thresholding the\nconsensus p-values. The theoretical analysis on the generalization error bound\nof the proposed classifier is provided and empirical studies on real data sets\nare conducted as well to demonstrate its effectiveness.\n", "date": "2024-10-16", "categories": []} {"id": "2410.09894", "title": "Inductive Conformal Prediction under Data Scarcity: Exploring the\n Impacts of Nonconformity Measures", "abstract": " Conformal prediction, which makes no distributional assumptions about the\ndata, has emerged as a powerful and reliable approach to uncertainty\nquantification in practical applications. The nonconformity measure used in\nconformal prediction quantifies how a test sample differs from the training\ndata and the effectiveness of a conformal prediction interval may depend\nheavily on the precise measure employed. The impact of this choice has,\nhowever, not been widely explored, especially when dealing with limited amounts\nof data. The primary objective of this study is to evaluate the performance of\nvarious nonconformity measures (absolute error-based, normalized absolute\nerror-based, and quantile-based measures) in terms of validity and efficiency\nwhen used in inductive conformal prediction. The focus is on small datasets,\nwhich is still a common setting in many real-world applications. Using\nsynthetic and real-world data, we assess how different characteristics -- such\nas dataset size, noise, and dimensionality -- can affect the efficiency of\nconformal prediction intervals. Our results show that although there are\ndifferences, no single nonconformity measure consistently outperforms the\nothers, as the effectiveness of each nonconformity measure is heavily\ninfluenced by the specific nature of the data. Additionally, we found that\nincreasing dataset size does not always improve efficiency, suggesting the\nimportance of fine-tuning models and, again, the need to carefully select the\nnonconformity measure for different applications.\n", "date": "2024-10-13", "categories": []} {"id": "2407.01794", "title": "Probabilistic Conformal Prediction with Approximate Conditional Validity", "abstract": " We develop a new method for generating prediction sets that combines the\nflexibility of conformal methods with an estimate of the conditional\ndistribution $P_{Y \\mid X}$. Existing methods, such as conformalized quantile\nregression and probabilistic conformal prediction, usually provide only a\nmarginal coverage guarantee. In contrast, our approach extends these frameworks\nto achieve approximately conditional coverage, which is crucial for many\npractical applications. Our prediction sets adapt to the behavior of the\npredictive distribution, making them effective even under high\nheteroscedasticity. While exact conditional guarantees are infeasible without\nassumptions on the underlying data distribution, we derive non-asymptotic\nbounds that depend on the total variation distance of the conditional\ndistribution and its estimate. Using extensive simulations, we show that our\nmethod consistently outperforms existing approaches in terms of conditional\ncoverage, leading to more reliable statistical inference in a variety of\napplications.\n", "date": "2024-07-01", "categories": []} {"id": "2408.13825", "title": "RoCP-GNN: Robust Conformal Prediction for Graph Neural Networks in\n Node-Classification", "abstract": " Graph Neural Networks (GNNs) have emerged as powerful tools for predicting\noutcomes in graph-structured data. However, a notable limitation of GNNs is\ntheir inability to provide robust uncertainty estimates, which undermines their\nreliability in contexts where errors are costly. One way to address this issue\nis by providing prediction sets that contain the true label with a predefined\nprobability margin. Our approach builds upon conformal prediction (CP), a\nframework that promises to construct statistically robust prediction sets or\nintervals. There are two primary challenges: first, given dependent data like\ngraphs, it is unclear whether the critical assumption in CP - exchangeability -\nstill holds when applied to node classification. Second, even if the\nexchangeability assumption is valid for conformalized link prediction, we need\nto ensure high efficiency, i.e., the resulting prediction set or the interval\nlength is small enough to provide useful information. In this article, we\npropose a novel approach termed Robust Conformal Prediction for GNNs\n(RoCP-GNN), which integrates conformal prediction (CP) directly into the GNN\ntraining process. This method generates prediction sets, instead of just point\npredictions, that are valid at a user-defined confidence level, assuming only\nexchangeability. Our approach robustly predicts outcomes with any predictive\nGNN model while quantifying the uncertainty in predictions within the realm of\ngraph-based semi-supervised learning (SSL). Experimental results demonstrate\nthat GNN models with size loss provide a statistically significant increase in\nperformance. We validate our approach on standard graph benchmark datasets by\ncoupling it with various state-of-the-art GNNs in node classification. The code\nwill be made available after publication.\n", "date": "2024-08-25", "categories": []} {"id": "2410.05444", "title": "Online scalable Gaussian processes with conformal prediction for\n guaranteed coverage", "abstract": " The Gaussian process (GP) is a Bayesian nonparametric paradigm that is widely\nadopted for uncertainty quantification (UQ) in a number of safety-critical\napplications, including robotics, healthcare, as well as surveillance. The\nconsistency of the resulting uncertainty values however, hinges on the premise\nthat the learning function conforms to the properties specified by the GP\nmodel, such as smoothness, periodicity and more, which may not be satisfied in\npractice, especially with data arriving on the fly. To combat against such\nmodel mis-specification, we propose to wed the GP with the prevailing conformal\nprediction (CP), a distribution-free post-processing framework that produces it\nprediction sets with a provably valid coverage under the sole assumption of\ndata exchangeability. However, this assumption is usually violated in the\nonline setting, where a prediction set is sought before revealing the true\nlabel. To ensure long-term coverage guarantee, we will adaptively set the key\nthreshold parameter based on the feedback whether the true label falls inside\nthe prediction set. Numerical results demonstrate the merits of the online\nGP-CP approach relative to existing alternatives in the long-term coverage\nperformance.\n", "date": "2024-10-07", "categories": []} {"id": "2410.02914", "title": "Streamlining Conformal Information Retrieval via Score Refinement", "abstract": " Information retrieval (IR) methods, like retrieval augmented generation, are\nfundamental to modern applications but often lack statistical guarantees.\nConformal prediction addresses this by retrieving sets guaranteed to include\nrelevant information, yet existing approaches produce large-sized sets,\nincurring high computational costs and slow response times. In this work, we\nintroduce a score refinement method that applies a simple monotone\ntransformation to retrieval scores, leading to significantly smaller conformal\nsets while maintaining their statistical guarantees. Experiments on various\nBEIR benchmarks validate the effectiveness of our approach in producing compact\nsets containing relevant information.\n", "date": "2024-10-03", "categories": []} {"id": "2409.17466", "title": "Adjusting Regression Models for Conditional Uncertainty Calibration", "abstract": " Conformal Prediction methods have finite-sample distribution-free marginal\ncoverage guarantees. However, they generally do not offer conditional coverage\nguarantees, which can be important for high-stakes decisions. In this paper, we\npropose a novel algorithm to train a regression function to improve the\nconditional coverage after applying the split conformal prediction procedure.\nWe establish an upper bound for the miscoverage gap between the conditional\ncoverage and the nominal coverage rate and propose an end-to-end algorithm to\ncontrol this upper bound. We demonstrate the efficacy of our method empirically\non synthetic and real-world datasets.\n", "date": "2024-09-26", "categories": []} {"id": "2409.14792", "title": "Adaptive Conformal Inference for Multi-Step Ahead Time-Series\n Forecasting Online", "abstract": " The aim of this paper is to propose an adaptation of the well known adaptive\nconformal inference (ACI) algorithm to achieve finite-sample coverage\nguarantees in multi-step ahead time-series forecasting in the online setting.\nACI dynamically adjusts significance levels, and comes with finite-sample\nguarantees on coverage, even for non-exchangeable data. Our multi-step ahead\nACI procedure inherits these guarantees at each prediction step, as well as for\nthe overall error rate. The multi-step ahead ACI algorithm can be used with\ndifferent target error and learning rates at different prediction steps, which\nis illustrated in our numerical examples, where we employ a version of the\nconfromalised ridge regression algorithm, adapted to multi-input multi-output\nforecasting. The examples serve to show how the method works in practice,\nillustrating the effect of variable target error and learning rates for\ndifferent prediction steps, which suggests that a balance may be struck between\nefficiency (interval width) and coverage.t\n", "date": "2024-09-23", "categories": []} {"id": "2409.13585", "title": "Neurosymbolic Conformal Classification", "abstract": " The last decades have seen a drastic improvement of Machine Learning (ML),\nmainly driven by Deep Learning (DL). However, despite the resounding successes\nof ML in many domains, the impossibility to provide guarantees of conformity\nand the fragility of ML systems (faced with distribution shifts, adversarial\nattacks, etc.) have prevented the design of trustworthy AI systems. Several\nresearch paths have been investigated to mitigate this fragility and provide\nsome guarantees regarding the behavior of ML systems, among which are\nneurosymbolic AI and conformal prediction. Neurosymbolic artificial\nintelligence is a growing field of research aiming to combine neural network\nlearning capabilities with the reasoning abilities of symbolic systems. One of\nthe objective of this hybridization can be to provide theoritical guarantees\nthat the output of the system will comply with some prior knowledge. Conformal\nprediction is a set of techniques that enable to take into account the\nuncertainty of ML systems by transforming the unique prediction into a set of\npredictions, called a confidence set. Interestingly, this comes with\nstatistical guarantees regarding the presence of the true label inside the\nconfidence set. Both approaches are distribution-free and model-agnostic. In\nthis paper, we see how these two approaches can complement one another. We\nintroduce several neurosymbolic conformal prediction techniques and explore\ntheir different characteristics (size of confidence sets, computational\ncomplexity, etc.).\n", "date": "2024-09-20", "categories": []} {"id": "2312.16160", "title": "SymmPI: Predictive Inference for Data with Group Symmetries", "abstract": " Quantifying the uncertainty of predictions is a core problem in modern\nstatistics. Methods for predictive inference have been developed under a\nvariety of assumptions, often -- for instance, in standard conformal prediction\n-- relying on the invariance of the distribution of the data under special\ngroups of transformations such as permutation groups. Moreover, many existing\nmethods for predictive inference aim to predict unobserved outcomes in\nsequences of feature-outcome observations. Meanwhile, there is interest in\npredictive inference under more general observation models (e.g., for partially\nobserved features) and for data satisfying more general distributional\nsymmetries (e.g., rotationally invariant or coordinate-independent observations\nin physics). Here we propose SymmPI, a methodology for predictive inference\nwhen data distributions have general group symmetries in arbitrary observation\nmodels. Our methods leverage the novel notion of distributional equivariant\ntransformations, which process the data while preserving their distributional\ninvariances. We show that SymmPI has valid coverage under distributional\ninvariance and characterize its performance under distribution shift,\nrecovering recent results as special cases. We apply SymmPI to predict\nunobserved values associated to vertices in a network, where the distribution\nis unchanged under relabelings that keep the network structure unchanged. In\nseveral simulations in a two-layer hierarchical model, and in an empirical data\nanalysis example, SymmPI performs favorably compared to existing methods.\n", "date": "2023-12-26", "categories": []} {"id": "2207.14219", "title": "A general framework for multi-step ahead adaptive conformal\n heteroscedastic time series forecasting", "abstract": " This paper introduces a novel model-agnostic algorithm called adaptive\nensemble batch multi-input multi-output conformalized quantile regression\n(AEnbMIMOCQR} that enables forecasters to generate multi-step ahead prediction\nintervals for a fixed pre-specified miscoverage rate in a distribution-free\nmanner. Our method is grounded on conformal prediction principles, however, it\ndoes not require data splitting and provides close to exact coverage even when\nthe data is not exchangeable. Moreover, the resulting prediction intervals,\nbesides being empirically valid along the forecast horizon, do not neglect\nheteroscedasticity. AEnbMIMOCQR is designed to be robust to distribution\nshifts, which means that its prediction intervals remain reliable over an\nunlimited period of time, without entailing retraining or imposing unrealistic\nstrict assumptions on the data-generating process. Through methodically\nexperimentation, we demonstrate that our approach outperforms other competitive\nmethods on both real-world and synthetic datasets. The code used in the\nexperimental part and a tutorial on how to use AEnbMIMOCQR can be found at the\nfollowing GitHub repository: https://github.com/Quilograma/AEnbMIMOCQR.\n", "date": "2022-07-28", "categories": []} {"id": "2310.15641", "title": "Guaranteed Coverage Prediction Intervals with Gaussian Process\n Regression", "abstract": " Gaussian Process Regression (GPR) is a popular regression method, which\nunlike most Machine Learning techniques, provides estimates of uncertainty for\nits predictions. These uncertainty estimates however, are based on the\nassumption that the model is well-specified, an assumption that is violated in\nmost practical applications, since the required knowledge is rarely available.\nAs a result, the produced uncertainty estimates can become very misleading; for\nexample the prediction intervals (PIs) produced for the 95% confidence level\nmay cover much less than 95% of the true labels. To address this issue, this\npaper introduces an extension of GPR based on a Machine Learning framework\ncalled, Conformal Prediction (CP). This extension guarantees the production of\nPIs with the required coverage even when the model is completely misspecified.\nThe proposed approach combines the advantages of GPR with the valid coverage\nguarantee of CP, while the performed experimental results demonstrate its\nsuperiority over existing methods.\n", "date": "2023-10-24", "categories": []} {"id": "2405.18723", "title": "Conformal Depression Prediction", "abstract": " While existing depression prediction methods based on deep learning show\npromise, their practical application is hindered by the lack of\ntrustworthiness, as these deep models are often deployed as black box models,\nleaving us uncertain on the confidence of their predictions. For high-risk\nclinical applications like depression prediction, uncertainty quantification is\nessential in decision-making. In this paper, we introduce conformal depression\nprediction (CDP), a depression prediction method with uncertainty\nquantification based on conformal prediction (CP), giving valid confidence\nintervals with theoretical coverage guarantees for the model predictions. CDP\nis a plug-and-play module that requires neither model retraining nor an\nassumption about the depression data distribution. As CDP provides only an\naverage coverage guarantee across all inputs rather than per-input performance\nguarantee, we further propose CDP-ACC, an improved conformal prediction with\napproximate conditional coverage. CDP-ACC firstly estimates the prediction\ndistribution through neighborhood relaxation, and then introduces a conformal\nscore function by constructing nested sequences, so as to provide a tighter\nprediction interval adaptive to specific input. We empirically demonstrate the\napplication of CDP in uncertainty-aware facial depression prediction, as well\nas the effectiveness and superiority of CDP-ACC on the AVEC 2013 and AVEC 2014\ndatasets. Our code is publicly available at https://github.com/PushineLee/CDP.\n", "date": "2024-05-29", "categories": []} {"id": "2408.05037", "title": "A conformalized learning of a prediction set with applications to\n medical imaging classification", "abstract": " Medical imaging classifiers can achieve high predictive accuracy, but\nquantifying their uncertainty remains an unresolved challenge, which prevents\ntheir deployment in medical clinics. We present an algorithm that can modify\nany classifier to produce a prediction set containing the true label with a\nuser-specified probability, such as 90%. We train a network to predict an\ninstance-based version of the Conformal Prediction threshold. The threshold is\nthen conformalized to ensure the required coverage. We applied the proposed\nalgorithm to several standard medical imaging classification datasets. The\nexperimental results demonstrate that our method outperforms current approaches\nin terms of smaller average size of the prediction set while maintaining the\ndesired coverage.\n", "date": "2024-08-09", "categories": []} {"id": "2403.07263", "title": "Adaptive Bounding Box Uncertainties via Two-Step Conformal Prediction", "abstract": " Quantifying a model's predictive uncertainty is essential for safety-critical\napplications such as autonomous driving. We consider quantifying such\nuncertainty for multi-object detection. In particular, we leverage conformal\nprediction to obtain uncertainty intervals with guaranteed coverage for object\nbounding boxes. One challenge in doing so is that bounding box predictions are\nconditioned on the object's class label. Thus, we develop a novel two-step\nconformal approach that propagates uncertainty in predicted class labels into\nthe uncertainty intervals of bounding boxes. This broadens the validity of our\nconformal coverage guarantees to include incorrectly classified objects, thus\noffering more actionable safety assurances. Moreover, we investigate novel\nensemble and quantile regression formulations to ensure the bounding box\nintervals are adaptive to object size, leading to a more balanced coverage.\nValidating our two-step approach on real-world datasets for 2D bounding box\nlocalization, we find that desired coverage levels are satisfied with\npractically tight predictive uncertainty intervals.\n", "date": "2024-03-12", "categories": []} {"id": "2406.10787", "title": "Evidential Uncertainty Sets in Deep Classifiers Using Conformal\n Prediction", "abstract": " In this paper, we propose Evidential Conformal Prediction (ECP) method for\nimage classifiers to generate the conformal prediction sets. Our method is\ndesigned based on a non-conformity score function that has its roots in\nEvidential Deep Learning (EDL) as a method of quantifying model (epistemic)\nuncertainty in DNN classifiers. We use evidence that are derived from the logit\nvalues of target labels to compute the components of our non-conformity score\nfunction: the heuristic notion of uncertainty in CP, uncertainty surprisal, and\nexpected utility. Our extensive experimental evaluation demonstrates that ECP\noutperforms three state-of-the-art methods for generating CP sets, in terms of\ntheir set sizes and adaptivity while maintaining the coverage of true labels.\n", "date": "2024-06-16", "categories": []} {"id": "2407.19938", "title": "Robust Conformal Volume Estimation in 3D Medical Images", "abstract": " Volumetry is one of the principal downstream applications of 3D medical image\nsegmentation, for example, to detect abnormal tissue growth or for surgery\nplanning. Conformal Prediction is a promising framework for uncertainty\nquantification, providing calibrated predictive intervals associated with\nautomatic volume measurements. However, this methodology is based on the\nhypothesis that calibration and test samples are exchangeable, an assumption\nthat is in practice often violated in medical image applications. A weighted\nformulation of Conformal Prediction can be framed to mitigate this issue, but\nits empirical investigation in the medical domain is still lacking. A potential\nreason is that it relies on the estimation of the density ratio between the\ncalibration and test distributions, which is likely to be intractable in\nscenarios involving high-dimensional data. To circumvent this, we propose an\nefficient approach for density ratio estimation relying on the compressed\nlatent representations generated by the segmentation model. Our experiments\ndemonstrate the efficiency of our approach to reduce the coverage error in the\npresence of covariate shifts, in both synthetic and real-world settings. Our\nimplementation is available at https://github.com/benolmbrt/wcp_miccai\n", "date": "2024-07-29", "categories": []} {"id": "2407.17377", "title": "Entropy Reweighted Conformal Classification", "abstract": " Conformal Prediction (CP) is a powerful framework for constructing prediction\nsets with guaranteed coverage. However, recent studies have shown that\nintegrating confidence calibration with CP can lead to a degradation in\nefficiency. In this paper, We propose an adaptive approach that considers the\nclassifier's uncertainty and employs entropy-based reweighting to enhance the\nefficiency of prediction sets for conformal classification. Our experimental\nresults demonstrate that this method significantly improves efficiency.\n", "date": "2024-07-24", "categories": []} {"id": "2407.15277", "title": "Conformal Predictions under Markovian Data", "abstract": " We study the split Conformal Prediction method when applied to Markovian\ndata. We quantify the gap in terms of coverage induced by the correlations in\nthe data (compared to exchangeable data). This gap strongly depends on the\nmixing properties of the underlying Markov chain, and we prove that it\ntypically scales as $\\sqrt{t_\\mathrm{mix}\\ln(n)/n}$ (where $t_\\mathrm{mix}$ is\nthe mixing time of the chain). We also derive upper bounds on the impact of the\ncorrelations on the size of the prediction set. Finally we present $K$-split\nCP, a method that consists in thinning the calibration dataset and that adapts\nto the mixing properties of the chain. Its coverage gap is reduced to\n$t_\\mathrm{mix}/(n\\ln(n))$ without really affecting the size of the prediction\nset. We finally test our algorithms on synthetic and real-world datasets.\n", "date": "2024-07-21", "categories": []} {"id": "2406.18763", "title": "Conformalized Link Prediction on Graph Neural Networks", "abstract": " Graph Neural Networks (GNNs) excel in diverse tasks, yet their applications\nin high-stakes domains are often hampered by unreliable predictions. Although\nnumerous uncertainty quantification methods have been proposed to address this\nlimitation, they often lack \\textit{rigorous} uncertainty estimates. This work\nmakes the first attempt to introduce a distribution-free and model-agnostic\nuncertainty quantification approach to construct a predictive interval with a\nstatistical guarantee for GNN-based link prediction. We term it as\n\\textit{conformalized link prediction.} Our approach builds upon conformal\nprediction (CP), a framework that promises to construct statistically robust\nprediction sets or intervals. We first theoretically and empirically establish\na permutation invariance condition for the application of CP in link prediction\ntasks, along with an exact test-time coverage. Leveraging the important\nstructural information in graphs, we then identify a novel and crucial\nconnection between a graph's adherence to the power law distribution and the\nefficiency of CP. This insight leads to the development of a simple yet\neffective sampling-based method to align the graph structure with a power law\ndistribution prior to the standard CP procedure. Extensive experiments\ndemonstrate that for conformalized link prediction, our approach achieves the\ndesired marginal coverage while significantly improving the efficiency of CP\ncompared to baseline methods.\n", "date": "2024-06-26", "categories": []} {"id": "2407.10976", "title": "Learning Cellular Network Connection Quality with Conformal", "abstract": " In this paper, we address the problem of uncertainty quantification for\ncellular network speed. It is a well-known fact that the actual internet speed\nexperienced by a mobile phone can fluctuate significantly, even when remaining\nin a single location. This high degree of variability underscores that mere\npoint estimation of network speed is insufficient. Rather, it is advantageous\nto establish a prediction interval that can encompass the expected range of\nspeed variations. In order to build an accurate network estimation map,\nnumerous mobile data need to be collected at different locations. Currently,\npublic datasets rely on users to upload data through apps. Although massive\ndata has been collected, the datasets suffer from significant noise due to the\nnature of cellular networks and various other factors. Additionally, the uneven\ndistribution of population density affects the spatial consistency of data\ncollection, leading to substantial uncertainty in the network quality maps\nderived from this data. We focus our analysis on large-scale internet-quality\ndatasets provided by Ookla to construct an estimated map of connection quality.\nTo improve the reliability of this map, we introduce a novel conformal\nprediction technique to build an uncertainty map. We identify regions with\nheightened uncertainty to prioritize targeted, manual data collection. In\naddition, the uncertainty map quantifies how reliable the prediction is in\ndifferent areas. Our method also leads to a sampling strategy that guides\nresearchers to selectively gather high-quality data that best complement the\ncurrent dataset to improve the overall accuracy of the prediction model.\n", "date": "2024-06-04", "categories": []} {"id": "2312.15799", "title": "Efficient Conformal Prediction under Data Heterogeneity", "abstract": " Conformal Prediction (CP) stands out as a robust framework for uncertainty\nquantification, which is crucial for ensuring the reliability of predictions.\nHowever, common CP methods heavily rely on data exchangeability, a condition\noften violated in practice. Existing approaches for tackling\nnon-exchangeability lead to methods that are not computable beyond the simplest\nexamples. This work introduces a new efficient approach to CP that produces\nprovably valid confidence sets for fairly general non-exchangeable data\ndistributions. We illustrate the general theory with applications to the\nchallenging setting of federated learning under data heterogeneity between\nagents. Our method allows constructing provably valid personalized prediction\nsets for agents in a fully federated way. The effectiveness of the proposed\nmethod is demonstrated in a series of experiments on real-world datasets.\n", "date": "2023-12-25", "categories": []} {"id": "2407.09165", "title": "Robust Yet Efficient Conformal Prediction Sets", "abstract": " Conformal prediction (CP) can convert any model's output into prediction sets\nguaranteed to include the true label with any user-specified probability.\nHowever, same as the model itself, CP is vulnerable to adversarial test\nexamples (evasion) and perturbed calibration data (poisoning). We derive\nprovably robust sets by bounding the worst-case change in conformity scores.\nOur tighter bounds lead to more efficient sets. We cover both continuous and\ndiscrete (sparse) data and our guarantees work both for evasion and poisoning\nattacks (on both features and labels).\n", "date": "2024-07-12", "categories": []} {"id": "2407.09173", "title": "Conformal Inductive Graph Neural Networks", "abstract": " Conformal prediction (CP) transforms any model's output into prediction sets\nguaranteed to include (cover) the true label. CP requires exchangeability, a\nrelaxation of the i.i.d. assumption, to obtain a valid distribution-free\ncoverage guarantee. This makes it directly applicable to transductive\nnode-classification. However, conventional CP cannot be applied in inductive\nsettings due to the implicit shift in the (calibration) scores caused by\nmessage passing with the new nodes. We fix this issue for both cases of node\nand edge-exchangeable graphs, recovering the standard coverage guarantee\nwithout sacrificing statistical efficiency. We further prove that the guarantee\nholds independently of the prediction time, e.g. upon arrival of a new\nnode/edge or at any subsequent moment.\n", "date": "2024-07-12", "categories": []} {"id": "2407.04407", "title": "Trustworthy Classification through Rank-Based Conformal Prediction Sets", "abstract": " Machine learning classification tasks often benefit from predicting a set of\npossible labels with confidence scores to capture uncertainty. However,\nexisting methods struggle with the high-dimensional nature of the data and the\nlack of well-calibrated probabilities from modern classification models. We\npropose a novel conformal prediction method that employs a rank-based score\nfunction suitable for classification models that predict the order of labels\ncorrectly, even if not well-calibrated. Our approach constructs prediction sets\nthat achieve the desired coverage rate while managing their size. We provide a\ntheoretical analysis of the expected size of the conformal prediction sets\nbased on the rank distribution of the underlying classifier. Through extensive\nexperiments, we demonstrate that our method outperforms existing techniques on\nvarious datasets, providing reliable uncertainty quantification. Our\ncontributions include a novel conformal prediction method, theoretical\nanalysis, and empirical evaluation. This work advances the practical deployment\nof machine learning systems by enabling reliable uncertainty quantification.\n", "date": "2024-07-05", "categories": []} {"id": "2406.03346", "title": "Normalizing Flows for Conformal Regression", "abstract": " Conformal Prediction (CP) algorithms estimate the uncertainty of a prediction\nmodel by calibrating its outputs on labeled data. The same calibration scheme\nusually applies to any model and data without modifications. The obtained\nprediction intervals are valid by construction but could be inefficient, i.e.\nunnecessarily big, if the prediction errors are not uniformly distributed over\nthe input space.\n We present a general scheme to localize the intervals by training the\ncalibration process. The standard prediction error is replaced by an optimized\ndistance metric that depends explicitly on the object attributes. Learning the\noptimal metric is equivalent to training a Normalizing Flow that acts on the\njoint distribution of the errors and the inputs. Unlike the Error Reweighting\nCP algorithm of Papadopoulos et al. (2008), the framework allows estimating the\ngap between nominal and empirical conditional validity. The approach is\ncompatible with existing locally-adaptive CP strategies based on re-weighting\nthe calibration samples and applies to any point-prediction model without\nretraining.\n", "date": "2024-06-05", "categories": []} {"id": "2210.17405", "title": "Exact and Approximate Conformal Inference for Multi-Output Regression", "abstract": " It is common in machine learning to estimate a response $y$ given covariate\ninformation $x$. However, these predictions alone do not quantify any\nuncertainty associated with said predictions. One way to overcome this\ndeficiency is with conformal inference methods, which construct a set\ncontaining the unobserved response $y$ with a prescribed probability.\nUnfortunately, even with a one-dimensional response, conformal inference is\ncomputationally expensive despite recent encouraging advances. In this paper,\nwe explore multi-output regression, delivering exact derivations of conformal\ninference $p$-values when the predictive model can be described as a linear\nfunction of $y$. Additionally, we propose \\texttt{unionCP} and a multivariate\nextension of \\texttt{rootCP} as efficient ways of approximating the conformal\nprediction region for a wide array of multi-output predictors, both linear and\nnonlinear, while preserving computational advantages. We also provide both\ntheoretical and empirical evidence of the effectiveness of these methods using\nboth real-world and simulated data.\n", "date": "2022-10-31", "categories": []} {"id": "2406.15760", "title": "ICM Ensemble with Novel Betting Functions for Concept Drift", "abstract": " This study builds upon our previous work by introducing a refined Inductive\nConformal Martingale (ICM) approach for addressing Concept Drift (CD).\nSpecifically, we enhance our previously proposed CAUTIOUS betting function to\nincorporate multiple density estimators for improving detection ability. We\nalso combine this betting function with two base estimators that have not been\npreviously utilized within the ICM framework: the Interpolated Histogram and\nNearest Neighbor Density Estimators. We assess these extensions using both a\nsingle ICM and an ensemble of ICMs. For the latter, we conduct a comprehensive\nexperimental investigation into the influence of the ensemble size on\nprediction accuracy and the number of available predictions. Our experimental\nresults on four benchmark datasets demonstrate that the proposed approach\nsurpasses our previous methodology in terms of performance while matching or in\nmany cases exceeding that of three contemporary state-of-the-art techniques.\n", "date": "2024-06-22", "categories": []} {"id": "2406.16766", "title": "Conformal time series decomposition with component-wise exchangeability", "abstract": " Conformal prediction offers a practical framework for distribution-free\nuncertainty quantification, providing finite-sample coverage guarantees under\nrelatively mild assumptions on data exchangeability. However, these assumptions\ncease to hold for time series due to their temporally correlated nature. In\nthis work, we present a novel use of conformal prediction for time series\nforecasting that incorporates time series decomposition. This approach allows\nus to model different temporal components individually. By applying specific\nconformal algorithms to each component and then merging the obtained prediction\nintervals, we customize our methods to account for the different\nexchangeability regimes underlying each component. Our decomposition-based\napproach is thoroughly discussed and empirically evaluated on synthetic and\nreal-world data. We find that the method provides promising results on\nwell-structured time series, but can be limited by factors such as the\ndecomposition step for more complex data.\n", "date": "2024-06-24", "categories": []} {"id": "2406.14483", "title": "Valid Error Bars for Neural Weather Models using Conformal Prediction", "abstract": " Neural weather models have shown immense potential as inexpensive and\naccurate alternatives to physics-based models. However, most models trained to\nperform weather forecasting do not quantify the uncertainty associated with\ntheir forecasts. This limits the trust in the model and the usefulness of the\nforecasts. In this work we construct and formalise a conformal prediction\nframework as a post-processing method for estimating this uncertainty. The\nmethod is model-agnostic and gives calibrated error bounds for all variables,\nlead times and spatial locations. No modifications are required to the model\nand the computational cost is negligible compared to model training. We\ndemonstrate the usefulness of the conformal prediction framework on a limited\narea neural weather model for the Nordic region. We further explore the\nadvantages of the framework for deterministic and probabilistic models.\n", "date": "2024-06-20", "categories": []} {"id": "2406.12262", "title": "Investigating Data Usage for Inductive Conformal Predictors", "abstract": " Inductive conformal predictors (ICPs) are algorithms that are able to\ngenerate prediction sets, instead of point predictions, which are valid at a\nuser-defined confidence level, only assuming exchangeability. These algorithms\nare useful for reliable machine learning and are increasing in popularity. The\nICP development process involves dividing development data into three parts:\ntraining, calibration and test. With access to limited or expensive development\ndata, it is an open question regarding the most efficient way to divide the\ndata. This study provides several experiments to explore this question and\nconsider the case for allowing overlap of examples between training and\ncalibration sets. Conclusions are drawn that will be of value to academics and\npractitioners planning to use ICPs.\n", "date": "2024-06-18", "categories": []} {"id": "2406.08884", "title": "The Penalized Inverse Probability Measure for Conformal Classification", "abstract": " The deployment of safe and trustworthy machine learning systems, and\nparticularly complex black box neural networks, in real-world applications\nrequires reliable and certified guarantees on their performance. The conformal\nprediction framework offers such formal guarantees by transforming any point\ninto a set predictor with valid, finite-set, guarantees on the coverage of the\ntrue at a chosen level of confidence. Central to this methodology is the notion\nof the nonconformity score function that assigns to each example a measure of\n''strangeness'' in comparison with the previously seen observations. While the\ncoverage guarantees are maintained regardless of the nonconformity measure, the\npoint predictor and the dataset, previous research has shown that the\nperformance of a conformal model, as measured by its efficiency (the average\nsize of the predicted sets) and its informativeness (the proportion of\nprediction sets that are singletons), is influenced by the choice of the\nnonconformity score function. The current work introduces the Penalized Inverse\nProbability (PIP) nonconformity score, and its regularized version RePIP, that\nallow the joint optimization of both efficiency and informativeness. Through\ntoy examples and empirical results on the task of crop and weed image\nclassification in agricultural robotics, the current work shows how PIP-based\nconformal classifiers exhibit precisely the desired behavior in comparison with\nother nonconformity measures and strike a good balance between informativeness\nand efficiency.\n", "date": "2024-06-13", "categories": []} {"id": "2401.13744", "title": "Conformal Prediction Sets Improve Human Decision Making", "abstract": " In response to everyday queries, humans explicitly signal uncertainty and\noffer alternative answers when they are unsure. Machine learning models that\noutput calibrated prediction sets through conformal prediction mimic this human\nbehaviour; larger sets signal greater uncertainty while providing alternatives.\nIn this work, we study the usefulness of conformal prediction sets as an aid\nfor human decision making by conducting a pre-registered randomized controlled\ntrial with conformal prediction sets provided to human subjects. With\nstatistical significance, we find that when humans are given conformal\nprediction sets their accuracy on tasks improves compared to fixed-size\nprediction sets with the same coverage guarantee. The results show that\nquantifying model uncertainty with conformal prediction is helpful for\nhuman-in-the-loop decision making and human-AI teams.\n", "date": "2024-01-24", "categories": []} {"id": "2404.02722", "title": "On-line conformalized neural networks ensembles for probabilistic\n forecasting of day-ahead electricity prices", "abstract": " Probabilistic electricity price forecasting (PEPF) is subject of increasing\ninterest, following the demand for proper quantification of prediction\nuncertainty, to support the operation in complex power markets with increasing\nshare of renewable generation. Distributional neural networks ensembles have\nbeen recently shown to outperform state of the art PEPF benchmarks. Still, they\nrequire critical reliability enhancements, as fail to pass the coverage tests\nat various steps on the prediction horizon. In this work, we propose a novel\napproach to PEPF, extending the state of the art neural networks ensembles\nbased methods through conformal inference based techniques, deployed within an\non-line recalibration procedure. Experiments have been conducted on multiple\nmarket regions, achieving day-ahead forecasts with improved hourly coverage and\nstable probabilistic scores.\n", "date": "2024-04-03", "categories": []} {"id": "2310.06430", "title": "Conformal Prediction for Deep Classifier via Label Ranking", "abstract": " Conformal prediction is a statistical framework that generates prediction\nsets containing ground-truth labels with a desired coverage guarantee. The\npredicted probabilities produced by machine learning models are generally\nmiscalibrated, leading to large prediction sets in conformal prediction. To\naddress this issue, we propose a novel algorithm named $\\textit{Sorted Adaptive\nPrediction Sets}$ (SAPS), which discards all the probability values except for\nthe maximum softmax probability. The key idea behind SAPS is to minimize the\ndependence of the non-conformity score on the probability values while\nretaining the uncertainty information. In this manner, SAPS can produce compact\nprediction sets and communicate instance-wise uncertainty. Extensive\nexperiments validate that SAPS not only lessens the prediction sets but also\nbroadly enhances the conditional coverage rate of prediction sets.\n", "date": "2023-10-10", "categories": []} {"id": "2401.04612", "title": "Distribution-Free Conformal Joint Prediction Regions for Neural Marked\n Temporal Point Processes", "abstract": " Sequences of labeled events observed at irregular intervals in continuous\ntime are ubiquitous across various fields. Temporal Point Processes (TPPs)\nprovide a mathematical framework for modeling these sequences, enabling\ninferences such as predicting the arrival time of future events and their\nassociated label, called mark. However, due to model misspecification or lack\nof training data, these probabilistic models may provide a poor approximation\nof the true, unknown underlying process, with prediction regions extracted from\nthem being unreliable estimates of the underlying uncertainty. This paper\ndevelops more reliable methods for uncertainty quantification in neural TPP\nmodels via the framework of conformal prediction. A primary objective is to\ngenerate a distribution-free joint prediction region for an event's arrival\ntime and mark, with a finite-sample marginal coverage guarantee. A key\nchallenge is to handle both a strictly positive, continuous response and a\ncategorical response, without distributional assumptions. We first consider a\nsimple but conservative approach that combines individual prediction regions\nfor the event's arrival time and mark. Then, we introduce a more effective\nmethod based on bivariate highest density regions derived from the joint\npredictive density of arrival times and marks. By leveraging the dependencies\nbetween these two variables, this method excludes unlikely combinations of the\ntwo, resulting in sharper prediction regions while still attaining the\npre-specified coverage level. We also explore the generation of individual\nunivariate prediction regions for events' arrival times and marks through\nconformal regression and classification techniques. Moreover, we evaluate the\nstronger notion of conditional coverage. Finally, through extensive\nexperimentation on both simulated and real-world datasets, we assess the\nvalidity and efficiency of these methods.\n", "date": "2024-01-09", "categories": []} {"id": "2405.06627", "title": "Conformal Validity Guarantees Exist for Any Data Distribution (and How\n to Find Them)", "abstract": " As artificial intelligence (AI) / machine learning (ML) gain widespread\nadoption, practitioners are increasingly seeking means to quantify and control\nthe risk these systems incur. This challenge is especially salient when such\nsystems have autonomy to collect their own data, such as in black-box\noptimization and active learning, where their actions induce sequential\nfeedback-loop shifts in the data distribution. Conformal prediction is a\npromising approach to uncertainty and risk quantification, but prior variants'\nvalidity guarantees have assumed some form of ``quasi-exchangeability'' on the\ndata distribution, thereby excluding many types of sequential shifts. In this\npaper we prove that conformal prediction can theoretically be extended to\n\\textit{any} joint data distribution, not just exchangeable or\nquasi-exchangeable ones. Although the most general case is exceedingly\nimpractical to compute, for concrete practical applications we outline a\nprocedure for deriving specific conformal algorithms for any data distribution,\nand we use this procedure to derive tractable algorithms for a series of\nAI/ML-agent-induced covariate shifts. We evaluate the proposed algorithms\nempirically on synthetic black-box optimization and active learning tasks.\n", "date": "2024-05-10", "categories": []} {"id": "2406.01960", "title": "Certifiably Byzantine-Robust Federated Conformal Prediction", "abstract": " Conformal prediction has shown impressive capacity in constructing\nstatistically rigorous prediction sets for machine learning models with\nexchangeable data samples. The siloed datasets, coupled with the escalating\nprivacy concerns related to local data sharing, have inspired recent\ninnovations extending conformal prediction into federated environments with\ndistributed data samples. However, this framework for distributed uncertainty\nquantification is susceptible to Byzantine failures. A minor subset of\nmalicious clients can significantly compromise the practicality of coverage\nguarantees. To address this vulnerability, we introduce a novel framework\nRob-FCP, which executes robust federated conformal prediction, effectively\ncountering malicious clients capable of reporting arbitrary statistics with the\nconformal calibration process. We theoretically provide the conformal coverage\nbound of Rob-FCP in the Byzantine setting and show that the coverage of Rob-FCP\nis asymptotically close to the desired coverage level. We also propose a\nmalicious client number estimator to tackle a more challenging setting where\nthe number of malicious clients is unknown to the defender and theoretically\nshows its effectiveness. We empirically demonstrate the robustness of Rob-FCP\nagainst diverse proportions of malicious clients under a variety of Byzantine\nattacks on five standard benchmark and real-world healthcare datasets.\n", "date": "2024-06-04", "categories": []} {"id": "2306.10193", "title": "Conformal Language Modeling", "abstract": " We propose a novel approach to conformal prediction for generative language\nmodels (LMs). Standard conformal prediction produces prediction sets -- in\nplace of single predictions -- that have rigorous, statistical performance\nguarantees. LM responses are typically sampled from the model's predicted\ndistribution over the large, combinatorial output space of natural language.\nTranslating this process to conformal prediction, we calibrate a stopping rule\nfor sampling different outputs from the LM that get added to a growing set of\ncandidates until we are confident that the output set is sufficient. Since some\nsamples may be low-quality, we also simultaneously calibrate and apply a\nrejection rule for removing candidates from the output set to reduce noise.\nSimilar to conformal prediction, we prove that the sampled set returned by our\nprocedure contains at least one acceptable answer with high probability, while\nstill being empirically precise (i.e., small) on average. Furthermore, within\nthis set of candidate responses, we show that we can also accurately identify\nsubsets of individual components -- such as phrases or sentences -- that are\neach independently correct (e.g., that are not \"hallucinations\"), again with\nstatistical guarantees. We demonstrate the promise of our approach on multiple\ntasks in open-domain question answering, text summarization, and radiology\nreport generation using different LM variants.\n", "date": "2023-06-16", "categories": []} {"id": "2311.05931", "title": "Early-Exit Neural Networks with Nested Prediction Sets", "abstract": " Early-exit neural networks (EENNs) enable adaptive and efficient inference by\nproviding predictions at multiple stages during the forward pass. In\nsafety-critical applications, these predictions are meaningful only when\naccompanied by reliable uncertainty estimates. A popular method for quantifying\nthe uncertainty of predictive models is the use of prediction sets. However, we\ndemonstrate that standard techniques such as conformal prediction and Bayesian\ncredible sets are not suitable for EENNs. They tend to generate non-nested sets\nacross exits, meaning that labels deemed improbable at one exit may reappear in\nthe prediction set of a subsequent exit. To address this issue, we investigate\nanytime-valid confidence sequences (AVCSs), an extension of traditional\nconfidence intervals tailored for data-streaming scenarios. These sequences are\ninherently nested and thus well-suited for an EENN's sequential predictions. We\nexplore the theoretical and practical challenges of using AVCSs in EENNs and\nshow that they indeed yield nested sets across exits. Thus our work presents a\npromising approach towards fast, yet still safe, predictive modeling\n", "date": "2023-11-10", "categories": []} {"id": "2405.07374", "title": "Conformalized Survival Distributions: A Generic Post-Process to Increase\n Calibration", "abstract": " Discrimination and calibration represent two important properties of survival\nanalysis, with the former assessing the model's ability to accurately rank\nsubjects and the latter evaluating the alignment of predicted outcomes with\nactual events. With their distinct nature, it is hard for survival models to\nsimultaneously optimize both of them especially as many previous results found\nimproving calibration tends to diminish discrimination performance. This paper\nintroduces a novel approach utilizing conformal regression that can improve a\nmodel's calibration without degrading discrimination. We provide theoretical\nguarantees for the above claim, and rigorously validate the efficiency of our\napproach across 11 real-world datasets, showcasing its practical applicability\nand robustness in diverse scenarios.\n", "date": "2024-05-12", "categories": []} {"id": "2406.01570", "title": "Single Trajectory Conformal Prediction", "abstract": " We study the performance of risk-controlling prediction sets (RCPS), an\nempirical risk minimization-based formulation of conformal prediction, with a\nsingle trajectory of temporally correlated data from an unknown stochastic\ndynamical system. First, we use the blocking technique to show that RCPS\nattains performance guarantees similar to those enjoyed in the iid setting\nwhenever data is generated by asymptotically stationary and contractive\ndynamics. Next, we use the decoupling technique to characterize the graceful\ndegradation in RCPS guarantees when the data generating process deviates from\nstationarity and contractivity. We conclude by discussing how these tools could\nbe used toward a unified analysis of online and offline conformal prediction\nalgorithms, which are currently treated with very different tools.\n", "date": "2024-06-03", "categories": []} {"id": "2405.18601", "title": "From Conformal Predictions to Confidence Regions", "abstract": " Conformal prediction methodologies have significantly advanced the\nquantification of uncertainties in predictive models. Yet, the construction of\nconfidence regions for model parameters presents a notable challenge, often\nnecessitating stringent assumptions regarding data distribution or merely\nproviding asymptotic guarantees. We introduce a novel approach termed CCR,\nwhich employs a combination of conformal prediction intervals for the model\noutputs to establish confidence regions for model parameters. We present\ncoverage guarantees under minimal assumptions on noise and that is valid in\nfinite sample regime. Our approach is applicable to both split conformal\npredictions and black-box methodologies including full or cross-conformal\napproaches. In the specific case of linear models, the derived confidence\nregion manifests as the feasible set of a Mixed-Integer Linear Program (MILP),\nfacilitating the deduction of confidence intervals for individual parameters\nand enabling robust optimization. We empirically compare CCR to recent\nadvancements in challenging settings such as with heteroskedastic and\nnon-Gaussian noise.\n", "date": "2024-05-28", "categories": []} {"id": "2402.01139", "title": "Online conformal prediction with decaying step sizes", "abstract": " We introduce a method for online conformal prediction with decaying step\nsizes. Like previous methods, ours possesses a retrospective guarantee of\ncoverage for arbitrary sequences. However, unlike previous methods, we can\nsimultaneously estimate a population quantile when it exists. Our theory and\nexperiments indicate substantially improved practical properties: in\nparticular, when the distribution is stable, the coverage is close to the\ndesired level for every time point, not just on average over the observed\nsequence.\n", "date": "2024-02-02", "categories": []} {"id": "2405.15912", "title": "Uncertainty Quantification for Neurosymbolic Programs via Compositional\n Conformal Prediction", "abstract": " Machine learning has become an effective tool for automatically annotating\nunstructured data (e.g., images) with structured labels (e.g., object\ndetections). As a result, a new programming paradigm called neurosymbolic\nprogramming has emerged where users write queries against these predicted\nannotations. However, due to the intrinsic fallibility of machine learning\nmodels, these programs currently lack any notion of correctness. In many\ndomains, users may want some kind of conservative guarantee that the results of\ntheir queries contain all possibly relevant instances. Conformal prediction has\nemerged as a promising strategy for quantifying uncertainty in machine learning\nby modifying models to predict sets of labels instead of individual labels; it\nprovides a probabilistic guarantee that the prediction set contains the true\nlabel with high probability. We propose a novel framework for adapting\nconformal prediction to neurosymbolic programs; our strategy is to represent\nprediction sets as abstract values in some abstract domain, and then to use\nabstract interpretation to propagate prediction sets through the program. Our\nstrategy satisfies three key desiderata: (i) correctness (i.e., the program\noutputs a prediction set that contains the true output with high probability),\n(ii) compositionality (i.e., we can quantify uncertainty separately for\ndifferent modules and then compose them together), and (iii) structured values\n(i.e., we can provide uncertainty quantification for structured values such as\nlists). When the full program is available ahead-of-time, we propose an\noptimization that incorporates conformal prediction at intermediate program\npoints to reduce imprecision in abstract interpretation. We evaluate our\napproach on programs that take MNIST and MS-COCO images as input, demonstrating\nthat it produces reasonably sized prediction sets while satisfying a coverage\nguarantee.\n", "date": "2024-05-24", "categories": []} {"id": "2405.15988", "title": "Transductive Confidence Machine and its application to Medical Data Sets", "abstract": " The Transductive Confidence Machine Nearest Neighbours (TCMNN) algorithm and\na supporting, simple user interface was developed. Different settings of the\nTCMNN algorithms' parameters were tested on medical data sets, in addition to\nthe use of different Minkowski metrics and polynomial kernels. The effect of\nincreasing the number of nearest neighbours and marking results with\nsignificance was also investigated. SVM implementation of the Transductive\nConfidence Machine was compared with Nearest Neighbours implementation. The\napplication of neural networks was investigated as a useful comparison to the\ntransductive algorithms.\n", "date": "2024-05-25", "categories": []} {"id": "2403.03850", "title": "Conformal prediction for multi-dimensional time series by ellipsoidal\n sets", "abstract": " Conformal prediction (CP) has been a popular method for uncertainty\nquantification because it is distribution-free, model-agnostic, and\ntheoretically sound. For forecasting problems in supervised learning, most CP\nmethods focus on building prediction intervals for univariate responses. In\nthis work, we develop a sequential CP method called $\\texttt{MultiDimSPCI}$\nthat builds prediction $\\textit{regions}$ for a multivariate response,\nespecially in the context of multivariate time series, which are not\nexchangeable. Theoretically, we estimate $\\textit{finite-sample}$\nhigh-probability bounds on the conditional coverage gap. Empirically, we\ndemonstrate that $\\texttt{MultiDimSPCI}$ maintains valid coverage on a wide\nrange of multivariate time series while producing smaller prediction regions\nthan CP and non-CP baselines.\n", "date": "2024-03-06", "categories": []} {"id": "2405.02648", "title": "A Conformal Prediction Score that is Robust to Label Noise", "abstract": " Conformal Prediction (CP) quantifies network uncertainty by building a small\nprediction set with a pre-defined probability that the correct class is within\nthis set. In this study we tackle the problem of CP calibration based on a\nvalidation set with noisy labels. We introduce a conformal score that is robust\nto label noise. The noise-free conformal score is estimated using the noisy\nlabeled data and the noise level. In the test phase the noise-free score is\nused to form the prediction set. We applied the proposed algorithm to several\nstandard medical imaging classification datasets. We show that our method\noutperforms current methods by a large margin, in terms of the average size of\nthe prediction set, while maintaining the required coverage.\n", "date": "2024-05-04", "categories": []} {"id": "2402.09623", "title": "Conformalized Adaptive Forecasting of Heterogeneous Trajectories", "abstract": " This paper presents a new conformal method for generating simultaneous\nforecasting bands guaranteed to cover the entire path of a new random\ntrajectory with sufficiently high probability. Prompted by the need for\ndependable uncertainty estimates in motion planning applications where the\nbehavior of diverse objects may be more or less unpredictable, we blend\ndifferent techniques from online conformal prediction of single and multiple\ntime series, as well as ideas for addressing heteroscedasticity in regression.\nThis solution is both principled, providing precise finite-sample guarantees,\nand effective, often leading to more informative predictions than prior\nmethods.\n", "date": "2024-02-14", "categories": []} {"id": "2405.04393", "title": "Efficient Online Set-valued Classification with Bandit Feedback", "abstract": " Conformal prediction is a distribution-free method that wraps a given machine\nlearning model and returns a set of plausible labels that contain the true\nlabel with a prescribed coverage rate. In practice, the empirical coverage\nachieved highly relies on fully observed label information from data both in\nthe training phase for model fitting and the calibration phase for quantile\nestimation. This dependency poses a challenge in the context of online learning\nwith bandit feedback, where a learner only has access to the correctness of\nactions (i.e., pulled an arm) but not the full information of the true label.\nIn particular, when the pulled arm is incorrect, the learner only knows that\nthe pulled one is not the true class label, but does not know which label is\ntrue. Additionally, bandit feedback further results in a smaller labeled\ndataset for calibration, limited to instances with correct actions, thereby\naffecting the accuracy of quantile estimation. To address these limitations, we\npropose Bandit Class-specific Conformal Prediction (BCCP), offering coverage\nguarantees on a class-specific granularity. Using an unbiased estimation of an\nestimand involving the true label, BCCP trains the model and makes set-valued\ninferences through stochastic gradient descent. Our approach overcomes the\nchallenges of sparsely labeled data in each iteration and generalizes the\nreliability and applicability of conformal prediction to online decision-making\nenvironments.\n", "date": "2024-05-07", "categories": []} {"id": "2405.02634", "title": "Onboard Out-of-Calibration Detection of Deep Learning Models using\n Conformal Prediction", "abstract": " The black box nature of deep learning models complicate their usage in\ncritical applications such as remote sensing. Conformal prediction is a method\nto ensure trust in such scenarios. Subject to data exchangeability, conformal\nprediction provides finite sample coverage guarantees in the form of a\nprediction set that is guaranteed to contain the true class within a user\ndefined error rate. In this letter we show that conformal prediction algorithms\nare related to the uncertainty of the deep learning model and that this\nrelation can be used to detect if the deep learning model is\nout-of-calibration. Popular classification models like Resnet50, Densenet161,\nInceptionV3, and MobileNetV2 are applied on remote sensing datasets such as the\nEuroSAT to demonstrate how under noisy scenarios the model outputs become\nuntrustworthy. Furthermore an out-of-calibration detection procedure relating\nthe model uncertainty and the average size of the conformal prediction set is\npresented.\n", "date": "2024-05-04", "categories": []} {"id": "2405.01976", "title": "Conformal Prediction for Natural Language Processing: A Survey", "abstract": " The rapid proliferation of large language models and natural language\nprocessing (NLP) applications creates a crucial need for uncertainty\nquantification to mitigate risks such as hallucinations and to enhance\ndecision-making reliability in critical applications. Conformal prediction is\nemerging as a theoretically sound and practically useful framework, combining\nflexibility with strong statistical guarantees. Its model-agnostic and\ndistribution-free nature makes it particularly promising to address the current\nshortcomings of NLP systems that stem from the absence of uncertainty\nquantification. This paper provides a comprehensive survey of conformal\nprediction techniques, their guarantees, and existing applications in NLP,\npointing to directions for future research and open challenges.\n", "date": "2024-05-03", "categories": []} {"id": "2405.02082", "title": "A comparative study of conformal prediction methods for valid\n uncertainty quantification in machine learning", "abstract": " In the past decades, most work in the area of data analysis and machine\nlearning was focused on optimizing predictive models and getting better results\nthan what was possible with existing models. To what extent the metrics with\nwhich such improvements were measured were accurately capturing the intended\ngoal, whether the numerical differences in the resulting values were\nsignificant, or whether uncertainty played a role in this study and if it\nshould have been taken into account, was of secondary importance. Whereas\nprobability theory, be it frequentist or Bayesian, used to be the gold standard\nin science before the advent of the supercomputer, it was quickly replaced in\nfavor of black box models and sheer computing power because of their ability to\nhandle large data sets. This evolution sadly happened at the expense of\ninterpretability and trustworthiness. However, while people are still trying to\nimprove the predictive power of their models, the community is starting to\nrealize that for many applications it is not so much the exact prediction that\nis of importance, but rather the variability or uncertainty.\n The work in this dissertation tries to further the quest for a world where\neveryone is aware of uncertainty, of how important it is and how to embrace it\ninstead of fearing it. A specific, though general, framework that allows anyone\nto obtain accurate uncertainty estimates is singled out and analysed. Certain\naspects and applications of the framework -- dubbed `conformal prediction' --\nare studied in detail. Whereas many approaches to uncertainty quantification\nmake strong assumptions about the data, conformal prediction is, at the time of\nwriting, the only framework that deserves the title `distribution-free'. No\nparametric assumptions have to be made and the nonparametric results also hold\nwithout having to resort to the law of large numbers in the asymptotic regime.\n", "date": "2024-05-03", "categories": []} {"id": "2310.05921", "title": "Conformal Decision Theory: Safe Autonomous Decisions from Imperfect\n Predictions", "abstract": " We introduce Conformal Decision Theory, a framework for producing safe\nautonomous decisions despite imperfect machine learning predictions. Examples\nof such decisions are ubiquitous, from robot planning algorithms that rely on\npedestrian predictions, to calibrating autonomous manufacturing to exhibit high\nthroughput and low error, to the choice of trusting a nominal policy versus\nswitching to a safe backup policy at run-time. The decisions produced by our\nalgorithms are safe in the sense that they come with provable statistical\nguarantees of having low risk without any assumptions on the world model\nwhatsoever; the observations need not be I.I.D. and can even be adversarial.\nThe theory extends results from conformal prediction to calibrate decisions\ndirectly, without requiring the construction of prediction sets. Experiments\ndemonstrate the utility of our approach in robot motion planning around humans,\nautomated stock trading, and robot manufacturing.\n", "date": "2023-10-09", "categories": []} {"id": "2302.07675", "title": "Guaranteed Dynamic Scheduling of Ultra-Reliable Low-Latency Traffic via\n Conformal Prediction", "abstract": " The dynamic scheduling of ultra-reliable and low-latency traffic (URLLC) in\nthe uplink can significantly enhance the efficiency of coexisting services,\nsuch as enhanced mobile broadband (eMBB) devices, by only allocating resources\nwhen necessary. The main challenge is posed by the uncertainty in the process\nof URLLC packet generation, which mandates the use of predictors for URLLC\ntraffic in the coming frames. In practice, such prediction may overestimate or\nunderestimate the amount of URLLC data to be generated, yielding either an\nexcessive or an insufficient amount of resources to be pre-emptively allocated\nfor URLLC packets. In this paper, we introduce a novel scheduler for URLLC\npackets that provides formal guarantees on reliability and latency irrespective\nof the quality of the URLLC traffic predictor. The proposed method leverages\nrecent advances in online conformal prediction (CP), and follows the principle\nof dynamically adjusting the amount of allocated resources so as to meet\nreliability and latency requirements set by the designer.\n", "date": "2023-02-15", "categories": []} {"id": "2401.11974", "title": "Cross-Validation Conformal Risk Control", "abstract": " Conformal risk control (CRC) is a recently proposed technique that applies\npost-hoc to a conventional point predictor to provide calibration guarantees.\nGeneralizing conformal prediction (CP), with CRC, calibration is ensured for a\nset predictor that is extracted from the point predictor to control a risk\nfunction such as the probability of miscoverage or the false negative rate. The\noriginal CRC requires the available data set to be split between training and\nvalidation data sets. This can be problematic when data availability is\nlimited, resulting in inefficient set predictors. In this paper, a novel CRC\nmethod is introduced that is based on cross-validation, rather than on\nvalidation as the original CRC. The proposed cross-validation CRC (CV-CRC)\nextends a version of the jackknife-minmax from CP to CRC, allowing for the\ncontrol of a broader range of risk functions. CV-CRC is proved to offer\ntheoretical guarantees on the average risk of the set predictor. Furthermore,\nnumerical experiments show that CV-CRC can reduce the average set size with\nrespect to CRC when the available data are limited.\n", "date": "2024-01-22", "categories": []} {"id": "2405.00417", "title": "Conformal Risk Control for Ordinal Classification", "abstract": " As a natural extension to the standard conformal prediction method, several\nconformal risk control methods have been recently developed and applied to\nvarious learning problems. In this work, we seek to control the conformal risk\nin expectation for ordinal classification tasks, which have broad applications\nto many real problems. For this purpose, we firstly formulated the ordinal\nclassification task in the conformal risk control framework, and provided\ntheoretic risk bounds of the risk control method. Then we proposed two types of\nloss functions specially designed for ordinal classification tasks, and\ndeveloped corresponding algorithms to determine the prediction set for each\ncase to control their risks at a desired level. We demonstrated the\neffectiveness of our proposed methods, and analyzed the difference between the\ntwo types of risks on three different datasets, including a simulated dataset,\nthe UTKFace dataset and the diabetic retinopathy detection dataset.\n", "date": "2024-05-01", "categories": []} {"id": "2309.08313", "title": "Conditional validity of heteroskedastic conformal regression", "abstract": " Conformal prediction, and split conformal prediction as a specific\nimplementation, offer a distribution-free approach to estimating prediction\nintervals with statistical guarantees. Recent work has shown that split\nconformal prediction can produce state-of-the-art prediction intervals when\nfocusing on marginal coverage, i.e. on a calibration dataset the method\nproduces on average prediction intervals that contain the ground truth with a\npredefined coverage level. However, such intervals are often not adaptive,\nwhich can be problematic for regression problems with heteroskedastic noise.\nThis paper tries to shed new light on how prediction intervals can be\nconstructed, using methods such as normalized and Mondrian conformal\nprediction, in such a way that they adapt to the heteroskedasticity of the\nunderlying process. Theoretical and experimental results are presented in which\nthese methods are compared in a systematic way. In particular, it is shown how\nthe conditional validity of a chosen conformal predictor can be related to\n(implicit) assumptions about the data-generating distribution.\n", "date": "2023-09-15", "categories": []} {"id": "2404.19651", "title": "Provably Robust Conformal Prediction with Improved Efficiency", "abstract": " Conformal prediction is a powerful tool to generate uncertainty sets with\nguaranteed coverage using any predictive model, under the assumption that the\ntraining and test data are i.i.d.. Recently, it has been shown that adversarial\nexamples are able to manipulate conformal methods to construct prediction sets\nwith invalid coverage rates, as the i.i.d. assumption is violated. To address\nthis issue, a recent work, Randomized Smoothed Conformal Prediction (RSCP), was\nfirst proposed to certify the robustness of conformal prediction methods to\nadversarial noise. However, RSCP has two major limitations: (i) its robustness\nguarantee is flawed when used in practice and (ii) it tends to produce large\nuncertainty sets. To address these limitations, we first propose a novel\nframework called RSCP+ to provide provable robustness guarantee in evaluation,\nwhich fixes the issues in the original RSCP method. Next, we propose two novel\nmethods, Post-Training Transformation (PTT) and Robust Conformal Training\n(RCT), to effectively reduce prediction set size with little computation\noverhead. Experimental results in CIFAR10, CIFAR100, and ImageNet suggest the\nbaseline method only yields trivial predictions including full label set, while\nour methods could boost the efficiency by up to $4.36\\times$, $5.46\\times$, and\n$16.9\\times$ respectively and provide practical robustness guarantee. Our codes\nare available at\nhttps://github.com/Trustworthy-ML-Lab/Provably-Robust-Conformal-Prediction.\n", "date": "2024-04-30", "categories": []} {"id": "2401.08876", "title": "Evaluating the Utility of Conformal Prediction Sets for AI-Advised Image\n Labeling", "abstract": " As deep neural networks are more commonly deployed in high-stakes domains,\ntheir black-box nature makes uncertainty quantification challenging. We\ninvestigate the presentation of conformal prediction sets--a distribution-free\nclass of methods for generating prediction sets with specified coverage--to\nexpress uncertainty in AI-advised decision-making. Through a large online\nexperiment, we compare the utility of conformal prediction sets to displays of\nTop-1 and Top-k predictions for AI-advised image labeling. In a pre-registered\nanalysis, we find that the utility of prediction sets for accuracy varies with\nthe difficulty of the task: while they result in accuracy on par with or less\nthan Top-1 and Top-k displays for easy images, prediction sets offer some\nadvantage in assisting humans in labeling out-of-distribution (OOD) images in\nthe setting that we studied, especially when the set size is small. Our results\nempirically pinpoint practical challenges of conformal prediction sets and\nprovide implications on how to incorporate them for real-world decision-making.\n", "date": "2024-01-16", "categories": []} {"id": "2404.17487", "title": "Conformal Prediction with Learned Features", "abstract": " In this paper, we focus on the problem of conformal prediction with\nconditional guarantees. Prior work has shown that it is impossible to construct\nnontrivial prediction sets with full conditional coverage guarantees. A wealth\nof research has considered relaxations of full conditional guarantees, relying\non some predefined uncertainty structures. Departing from this line of\nthinking, we propose Partition Learning Conformal Prediction (PLCP), a\nframework to improve conditional validity of prediction sets through learning\nuncertainty-guided features from the calibration data. We implement PLCP\nefficiently with alternating gradient descent, utilizing off-the-shelf machine\nlearning models. We further analyze PLCP theoretically and provide conditional\nguarantees for infinite and finite sample sizes. Finally, our experimental\nresults over four real-world and synthetic datasets show the superior\nperformance of PLCP compared to state-of-the-art methods in terms of coverage\nand length in both classification and regression scenarios.\n", "date": "2024-04-26", "categories": []} {"id": "2404.13731", "title": "Training-Conditional Coverage Bounds for Uniformly Stable Learning\n Algorithms", "abstract": " The training-conditional coverage performance of the conformal prediction is\nknown to be empirically sound. Recently, there have been efforts to support\nthis observation with theoretical guarantees. The training-conditional coverage\nbounds for jackknife+ and full-conformal prediction regions have been\nestablished via the notion of $(m,n)$-stability by Liang and Barber~[2023].\nAlthough this notion is weaker than uniform stability, it is not clear how to\nevaluate it for practical models. In this paper, we study the\ntraining-conditional coverage bounds of full-conformal, jackknife+, and CV+\nprediction regions from a uniform stability perspective which is known to hold\nfor empirical risk minimization over reproducing kernel Hilbert spaces with\nconvex regularization. We derive coverage bounds for finite-dimensional models\nby a concentration argument for the (estimated) predictor function, and compare\nthe bounds with existing ones under ridge regression.\n", "date": "2024-04-21", "categories": []} {"id": "2404.08168", "title": "Conformal Prediction via Regression-as-Classification", "abstract": " Conformal prediction (CP) for regression can be challenging, especially when\nthe output distribution is heteroscedastic, multimodal, or skewed. Some of the\nissues can be addressed by estimating a distribution over the output, but in\nreality, such approaches can be sensitive to estimation error and yield\nunstable intervals.~Here, we circumvent the challenges by converting regression\nto a classification problem and then use CP for classification to obtain CP\nsets for regression.~To preserve the ordering of the continuous-output space,\nwe design a new loss function and make necessary modifications to the CP\nclassification techniques.~Empirical results on many benchmarks shows that this\nsimple approach gives surprisingly good results on many practical problems.\n", "date": "2024-04-12", "categories": []} {"id": "2403.01216", "title": "API Is Enough: Conformal Prediction for Large Language Models Without\n Logit-Access", "abstract": " This study aims to address the pervasive challenge of quantifying uncertainty\nin large language models (LLMs) without logit-access. Conformal Prediction\n(CP), known for its model-agnostic and distribution-free features, is a desired\napproach for various LLMs and data distributions. However, existing CP methods\nfor LLMs typically assume access to the logits, which are unavailable for some\nAPI-only LLMs. In addition, logits are known to be miscalibrated, potentially\nleading to degraded CP performance. To tackle these challenges, we introduce a\nnovel CP method that (1) is tailored for API-only LLMs without logit-access;\n(2) minimizes the size of prediction sets; and (3) ensures a statistical\nguarantee of the user-defined coverage. The core idea of this approach is to\nformulate nonconformity measures using both coarse-grained (i.e., sample\nfrequency) and fine-grained uncertainty notions (e.g., semantic similarity).\nExperimental results on both close-ended and open-ended Question Answering\ntasks show our approach can mostly outperform the logit-based CP baselines.\n", "date": "2024-03-02", "categories": []} {"id": "2211.16462", "title": "Will My Robot Achieve My Goals? Predicting the Probability that an MDP\n Policy Reaches a User-Specified Behavior Target", "abstract": " As an autonomous system performs a task, it should maintain a calibrated\nestimate of the probability that it will achieve the user's goal. If that\nprobability falls below some desired level, it should alert the user so that\nappropriate interventions can be made. This paper considers settings where the\nuser's goal is specified as a target interval for a real-valued performance\nsummary, such as the cumulative reward, measured at a fixed horizon $H$. At\neach time $t \\in \\{0, \\ldots, H-1\\}$, our method produces a calibrated estimate\nof the probability that the final cumulative reward will fall within a\nuser-specified target interval $[y^-,y^+].$ Using this estimate, the autonomous\nsystem can raise an alarm if the probability drops below a specified threshold.\nWe compute the probability estimates by inverting conformal prediction. Our\nstarting point is the Conformalized Quantile Regression (CQR) method of Romano\net al., which applies split-conformal prediction to the results of quantile\nregression. CQR is not invertible, but by using the conditional cumulative\ndistribution function (CDF) as the non-conformity measure, we show how to\nobtain an invertible modification that we call Probability-space Conformalized\nQuantile Regression (PCQR). Like CQR, PCQR produces well-calibrated conditional\nprediction intervals with finite-sample marginal guarantees. By inverting PCQR,\nwe obtain guarantees for the probability that the cumulative reward of an\nautonomous system will fall below a threshold sampled from the marginal\ndistribution of the response variable (i.e., a calibrated CDF estimate) that we\nemploy to predict coverage probabilities for user-specified target intervals.\nExperiments on two domains confirm that these probabilities are\nwell-calibrated.\n", "date": "2022-11-29", "categories": []} {"id": "2403.19950", "title": "Coverage-Guaranteed Prediction Sets for Out-of-Distribution Data", "abstract": " Out-of-distribution (OOD) generalization has attracted increasing research\nattention in recent years, due to its promising experimental results in\nreal-world applications. In this paper,we study the confidence set prediction\nproblem in the OOD generalization setting. Split conformal prediction (SCP) is\nan efficient framework for handling the confidence set prediction problem.\nHowever, the validity of SCP requires the examples to be exchangeable, which is\nviolated in the OOD setting. Empirically, we show that trivially applying SCP\nresults in a failure to maintain the marginal coverage when the unseen target\ndomain is different from the source domain. To address this issue, we develop a\nmethod for forming confident prediction sets in the OOD setting and\ntheoretically prove the validity of our method. Finally, we conduct experiments\non simulated data to empirically verify the correctness of our theory and the\nvalidity of our proposed method.\n", "date": "2024-03-29", "categories": []} {"id": "2403.18973", "title": "Conformal Intent Classification and Clarification for Fast and Accurate\n Intent Recognition", "abstract": " We present Conformal Intent Classification and Clarification (CICC), a\nframework for fast and accurate intent classification for task-oriented\ndialogue systems. The framework turns heuristic uncertainty scores of any\nintent classifier into a clarification question that is guaranteed to contain\nthe true intent at a pre-defined confidence level. By disambiguating between a\nsmall number of likely intents, the user query can be resolved quickly and\naccurately. Additionally, we propose to augment the framework for out-of-scope\ndetection. In a comparative evaluation using seven intent recognition datasets\nwe find that CICC generates small clarification questions and is capable of\nout-of-scope detection. CICC can help practitioners and researchers\nsubstantially in improving the user experience of dialogue agents with specific\nclarification questions.\n", "date": "2024-03-27", "categories": []} {"id": "2403.19082", "title": "Enhancing Conformal Prediction Using E-Test Statistics", "abstract": " Conformal Prediction (CP) serves as a robust framework that quantifies\nuncertainty in predictions made by Machine Learning (ML) models. Unlike\ntraditional point predictors, CP generates statistically valid prediction\nregions, also known as prediction intervals, based on the assumption of data\nexchangeability. Typically, the construction of conformal predictions hinges on\np-values. This paper, however, ventures down an alternative path, harnessing\nthe power of e-test statistics to augment the efficacy of conformal predictions\nby introducing a BB-predictor (bounded from the below predictor).\n", "date": "2024-03-28", "categories": []} {"id": "2403.15025", "title": "Robust Conformal Prediction under Distribution Shift via\n Physics-Informed Structural Causal Model", "abstract": " Uncertainty is critical to reliable decision-making with machine learning.\nConformal prediction (CP) handles uncertainty by predicting a set on a test\ninput, hoping the set to cover the true label with at least $(1-\\alpha)$\nconfidence. This coverage can be guaranteed on test data even if the marginal\ndistributions $P_X$ differ between calibration and test datasets. However, as\nit is common in practice, when the conditional distribution $P_{Y|X}$ is\ndifferent on calibration and test data, the coverage is not guaranteed and it\nis essential to measure and minimize the coverage loss under distributional\nshift at \\textit{all} possible confidence levels. To address these issues, we\nupper bound the coverage difference at all levels using the cumulative density\nfunctions of calibration and test conformal scores and Wasserstein distance.\nInspired by the invariance of physics across data distributions, we propose a\nphysics-informed structural causal model (PI-SCM) to reduce the upper bound. We\nvalidated that PI-SCM can improve coverage robustness along confidence level\nand test domain on a traffic speed prediction task and an epidemic spread task\nwith multiple real-world datasets.\n", "date": "2024-03-22", "categories": []} {"id": "2212.03281", "title": "Copula Conformal Prediction for Multi-step Time Series Forecasting", "abstract": " Accurate uncertainty measurement is a key step to building robust and\nreliable machine learning systems. Conformal prediction is a distribution-free\nuncertainty quantification algorithm popular for its ease of implementation,\nstatistical coverage guarantees, and versatility for underlying forecasters.\nHowever, existing conformal prediction algorithms for time series are limited\nto single-step prediction without considering the temporal dependency. In this\npaper, we propose a Copula Conformal Prediction algorithm for multivariate,\nmulti-step Time Series forecasting, CopulaCPTS. We prove that CopulaCPTS has\nfinite sample validity guarantee. On several synthetic and real-world\nmultivariate time series datasets, we show that CopulaCPTS produces more\ncalibrated and sharp confidence intervals for multi-step prediction tasks than\nexisting techniques.\n", "date": "2022-12-06", "categories": []} {"id": "2310.18108", "title": "Transductive conformal inference with adaptive scores", "abstract": " Conformal inference is a fundamental and versatile tool that provides\ndistribution-free guarantees for many machine learning tasks. We consider the\ntransductive setting, where decisions are made on a test sample of $m$ new\npoints, giving rise to $m$ conformal $p$-values. While classical results only\nconcern their marginal distribution, we show that their joint distribution\nfollows a P\\'olya urn model, and establish a concentration inequality for their\nempirical distribution function. The results hold for arbitrary exchangeable\nscores, including adaptive ones that can use the covariates of the\ntest+calibration samples at training stage for increased accuracy. We\ndemonstrate the usefulness of these theoretical results through uniform,\nin-probability guarantees for two machine learning tasks of current interest:\ninterval prediction for transductive transfer learning and novelty detection\nbased on two-class classification.\n", "date": "2023-10-27", "categories": []} {"id": "2312.07460", "title": "Empirical Validation of Conformal Prediction for Trustworthy Skin\n Lesions Classification", "abstract": " Background and objective: Uncertainty quantification is a pivotal field that\ncontributes to realizing reliable and robust systems. It becomes instrumental\nin fortifying safe decisions by providing complementary information,\nparticularly within high-risk applications. existing studies have explored\nvarious methods that often operate under specific assumptions or necessitate\nsubstantial modifications to the network architecture to effectively account\nfor uncertainties. The objective of this paper is to study Conformal\nPrediction, an emerging distribution-free uncertainty quantification technique,\nand provide a comprehensive understanding of the advantages and limitations\ninherent in various methods within the medical imaging field.\n Methods: In this study, we developed Conformal Prediction, Monte Carlo\nDropout, and Evidential Deep Learning approaches to assess uncertainty\nquantification in deep neural networks. The effectiveness of these methods is\nevaluated using three public medical imaging datasets focused on detecting\npigmented skin lesions and blood cell types.\n Results: The experimental results demonstrate a significant enhancement in\nuncertainty quantification with the utilization of the Conformal Prediction\nmethod, surpassing the performance of the other two methods. Furthermore, the\nresults present insights into the effectiveness of each uncertainty method in\nhandling Out-of-Distribution samples from domain-shifted datasets. Our code is\navailable at:\n Conclusions: Our conclusion highlights a robust and consistent performance of\nconformal prediction across diverse testing conditions. This positions it as\nthe preferred choice for decision-making in safety-critical applications.\n", "date": "2023-12-12", "categories": []} {"id": "2403.11348", "title": "COLEP: Certifiably Robust Learning-Reasoning Conformal Prediction via\n Probabilistic Circuits", "abstract": " Conformal prediction has shown spurring performance in constructing\nstatistically rigorous prediction sets for arbitrary black-box machine learning\nmodels, assuming the data is exchangeable. However, even small adversarial\nperturbations during the inference can violate the exchangeability assumption,\nchallenge the coverage guarantees, and result in a subsequent decline in\nempirical coverage. In this work, we propose a certifiably robust\nlearning-reasoning conformal prediction framework (COLEP) via probabilistic\ncircuits, which comprise a data-driven learning component that trains\nstatistical models to learn different semantic concepts, and a reasoning\ncomponent that encodes knowledge and characterizes the relationships among the\ntrained models for logic reasoning. To achieve exact and efficient reasoning,\nwe employ probabilistic circuits (PCs) within the reasoning component.\nTheoretically, we provide end-to-end certification of prediction coverage for\nCOLEP in the presence of bounded adversarial perturbations. We also provide\ncertified coverage considering the finite size of the calibration set.\nFurthermore, we prove that COLEP achieves higher prediction coverage and\naccuracy over a single model as long as the utilities of knowledge models are\nnon-trivial. Empirically, we show the validity and tightness of our certified\ncoverage, demonstrating the robust conformal prediction of COLEP on various\ndatasets, including GTSRB, CIFAR10, and AwA2. We show that COLEP achieves up to\n12% improvement in certified coverage on GTSRB, 9% on CIFAR-10, and 14% on\nAwA2.\n", "date": "2024-03-17", "categories": []} {"id": "2403.10368", "title": "Conformal Predictions for Probabilistically Robust Scalable Machine\n Learning Classification", "abstract": " Conformal predictions make it possible to define reliable and robust learning\nalgorithms. But they are essentially a method for evaluating whether an\nalgorithm is good enough to be used in practice. To define a reliable learning\nframework for classification from the very beginning of its design, the concept\nof scalable classifier was introduced to generalize the concept of classical\nclassifier by linking it to statistical order theory and probabilistic learning\ntheory. In this paper, we analyze the similarities between scalable classifiers\nand conformal predictions by introducing a new definition of a score function\nand defining a special set of input variables, the conformal safety set, which\ncan identify patterns in the input space that satisfy the error coverage\nguarantee, i.e., that the probability of observing the wrong (possibly unsafe)\nlabel for points belonging to this set is bounded by a predefined $\\varepsilon$\nerror level. We demonstrate the practical implications of this framework\nthrough an application in cybersecurity for identifying DNS tunneling attacks.\nOur work contributes to the development of probabilistically robust and\nreliable machine learning models.\n", "date": "2024-03-15", "categories": []} {"id": "2306.07254", "title": "On the Expected Size of Conformal Prediction Sets", "abstract": " While conformal predictors reap the benefits of rigorous statistical\nguarantees on their error frequency, the size of their corresponding prediction\nsets is critical to their practical utility. Unfortunately, there is currently\na lack of finite-sample analysis and guarantees for their prediction set sizes.\nTo address this shortfall, we theoretically quantify the expected size of the\nprediction sets under the split conformal prediction framework. As this precise\nformulation cannot usually be calculated directly, we further derive point\nestimates and high-probability interval bounds that can be empirically\ncomputed, providing a practical method for characterizing the expected set\nsize. We corroborate the efficacy of our results with experiments on real-world\ndatasets for both regression and classification problems.\n", "date": "2023-06-12", "categories": []} {"id": "2311.12688", "title": "On the Out-of-Distribution Coverage of Combining Split Conformal\n Prediction and Bayesian Deep Learning", "abstract": " Bayesian deep learning and conformal prediction are two methods that have\nbeen used to convey uncertainty and increase safety in machine learning\nsystems. We focus on combining Bayesian deep learning with split conformal\nprediction and how this combination effects out-of-distribution coverage;\nparticularly in the case of multiclass image classification. We suggest that if\nthe model is generally underconfident on the calibration set, then the\nresultant conformal sets may exhibit worse out-of-distribution coverage\ncompared to simple predictive credible sets. Conversely, if the model is\noverconfident on the calibration set, the use of conformal prediction may\nimprove out-of-distribution coverage. We evaluate prediction sets as a result\nof combining split conformal methods and neural networks trained with (i)\nstochastic gradient descent, (ii) deep ensembles, and (iii) mean-field\nvariational inference. Our results suggest that combining Bayesian deep\nlearning models with split conformal prediction can, in some cases, cause\nunintended consequences such as reducing out-of-distribution coverage.\n", "date": "2023-11-21", "categories": []} {"id": "2302.02237", "title": "Conformalized Semi-supervised Random Forest for Classification and\n Abnormality Detection", "abstract": " The Random Forests classifier, a widely utilized off-the-shelf classification\ntool, assumes training and test samples come from the same distribution as\nother standard classifiers. However, in safety-critical scenarios like medical\ndiagnosis and network attack detection, discrepancies between the training and\ntest sets, including the potential presence of novel outlier samples not\nappearing during training, can pose significant challenges. To address this\nproblem, we introduce the Conformalized Semi-Supervised Random Forest\n(CSForest), which couples the conformalization technique Jackknife+aB with\nsemi-supervised tree ensembles to construct a set-valued prediction $C(x)$.\nInstead of optimizing over the training distribution, CSForest employs\nunlabeled test samples to enhance accuracy and flag unseen outliers by\ngenerating an empty set. Theoretically, we establish CSForest to cover true\nlabels for previously observed inlier classes under arbitrarily label-shift in\nthe test data. We compare CSForest with state-of-the-art methods using\nsynthetic examples and various real-world datasets, under different types of\ndistribution changes in the test domain. Our results highlight CSForest's\neffective prediction of inliers and its ability to detect outlier samples\nunique to the test data. In addition, CSForest shows persistently good\nperformance as the sizes of the training and test sets vary. Codes of CSForest\nare available at https://github.com/yujinhan98/CSForest.\n", "date": "2023-02-04", "categories": []} {"id": "2309.05092", "title": "Adaptive conformal classification with noisy labels", "abstract": " This paper develops novel conformal prediction methods for classification\ntasks that can automatically adapt to random label contamination in the\ncalibration sample, leading to more informative prediction sets with stronger\ncoverage guarantees compared to state-of-the-art approaches. This is made\npossible by a precise characterization of the effective coverage inflation (or\ndeflation) suffered by standard conformal inferences in the presence of label\ncontamination, which is then made actionable through new calibration\nalgorithms. Our solution is flexible and can leverage different modeling\nassumptions about the label contamination process, while requiring no knowledge\nof the underlying data distribution or of the inner workings of the\nmachine-learning classifier. The advantages of the proposed methods are\ndemonstrated through extensive simulations and an application to object\nclassification with the CIFAR-10H image data set.\n", "date": "2023-09-10", "categories": []} {"id": "2402.10978", "title": "Language Models with Conformal Factuality Guarantees", "abstract": " Guaranteeing the correctness and factuality of language model (LM) outputs is\na major open problem. In this work, we propose conformal factuality, a\nframework that can ensure high probability correctness guarantees for LMs by\nconnecting language modeling and conformal prediction. We observe that the\ncorrectness of an LM output is equivalent to an uncertainty quantification\nproblem, where the uncertainty sets are defined as the entailment set of an\nLM's output. Using this connection, we show that conformal prediction in\nlanguage models corresponds to a back-off algorithm that provides high\nprobability correctness guarantees by progressively making LM outputs less\nspecific (and expanding the associated uncertainty sets). This approach applies\nto any black-box LM and requires very few human-annotated samples. Evaluations\nof our approach on closed book QA (FActScore, NaturalQuestions) and reasoning\ntasks (MATH) show that our approach can provide 80-90% correctness guarantees\nwhile retaining the majority of the LM's original output.\n", "date": "2024-02-15", "categories": []} {"id": "2402.12307", "title": "Multi-View Conformal Learning for Heterogeneous Sensor Fusion", "abstract": " Being able to assess the confidence of individual predictions in machine\nlearning models is crucial for decision making scenarios. Specially, in\ncritical applications such as medical diagnosis, security, and unmanned\nvehicles, to name a few. In the last years, complex predictive models have had\ngreat success in solving hard tasks and new methods are being proposed every\nday. While the majority of new developments in machine learning models focus on\nimproving the overall performance, less effort is put on assessing the\ntrustworthiness of individual predictions, and even to a lesser extent, in the\ncontext of sensor fusion. To this end, we build and test multi-view and\nsingle-view conformal models for heterogeneous sensor fusion. Our models\nprovide theoretical marginal confidence guarantees since they are based on the\nconformal prediction framework. We also propose a multi-view semi-conformal\nmodel based on sets intersection. Through comprehensive experimentation, we\nshow that multi-view models perform better than single-view models not only in\nterms of accuracy-based performance metrics (as it has already been shown in\nseveral previous works) but also in conformal measures that provide uncertainty\nestimation. Our results also showed that multi-view models generate prediction\nsets with less uncertainty compared to single-view models.\n", "date": "2024-02-19", "categories": []} {"id": "2402.10723", "title": "Conformalized Credal Set Predictors", "abstract": " Credal sets are sets of probability distributions that are considered as\ncandidates for an imprecisely known ground-truth distribution. In machine\nlearning, they have recently attracted attention as an appealing formalism for\nuncertainty representation, in particular due to their ability to represent\nboth the aleatoric and epistemic uncertainty in a prediction. However, the\ndesign of methods for learning credal set predictors remains a challenging\nproblem. In this paper, we make use of conformal prediction for this purpose.\nMore specifically, we propose a method for predicting credal sets in the\nclassification task, given training data labeled by probability distributions.\nSince our method inherits the coverage guarantees of conformal prediction, our\nconformal credal sets are guaranteed to be valid with high probability (without\nany assumptions on model or distribution). We demonstrate the applicability of\nour method to natural language inference, a highly ambiguous natural language\ntask where it is common to obtain multiple annotations per example.\n", "date": "2024-02-16", "categories": []} {"id": "2402.07357", "title": "Regression Trees for Fast and Adaptive Prediction Intervals", "abstract": " Predictive models make mistakes. Hence, there is a need to quantify the\nuncertainty associated with their predictions. Conformal inference has emerged\nas a powerful tool to create statistically valid prediction regions around\npoint predictions, but its naive application to regression problems yields\nnon-adaptive regions. New conformal scores, often relying upon quantile\nregressors or conditional density estimators, aim to address this limitation.\nAlthough they are useful for creating prediction bands, these scores are\ndetached from the original goal of quantifying the uncertainty around an\narbitrary predictive model. This paper presents a new, model-agnostic family of\nmethods to calibrate prediction intervals for regression problems with local\ncoverage guarantees. Our approach is based on pursuing the coarsest partition\nof the feature space that approximates conditional coverage. We create this\npartition by training regression trees and Random Forests on conformity scores.\nOur proposal is versatile, as it applies to various conformity scores and\nprediction settings and demonstrates superior scalability and performance\ncompared to established baselines in simulated and real-world datasets. We\nprovide a Python package clover that implements our methods using the standard\nscikit-learn interface.\n", "date": "2024-02-12", "categories": []} {"id": "2402.00707", "title": "Non-Exchangeable Conformal Language Generation with Nearest Neighbors", "abstract": " Quantifying uncertainty in automatically generated text is important for\nletting humans check potential hallucinations and making systems more reliable.\nConformal prediction is an attractive framework to provide predictions imbued\nwith statistical guarantees, however, its application to text generation is\nchallenging since any i.i.d. assumptions are not realistic. In this paper, we\nbridge this gap by leveraging recent results on non-exchangeable conformal\nprediction, which still ensures bounds on coverage. The result,\nnon-exchangeable conformal nucleus sampling, is a novel extension of the\nconformal prediction framework to generation based on nearest neighbors. Our\nmethod can be used post-hoc for an arbitrary model without extra training and\nsupplies token-level, calibrated prediction sets equipped with statistical\nguarantees. Experiments in machine translation and language modeling show\nencouraging results in generation quality. By also producing tighter prediction\nsets with good coverage, we thus give a more theoretically principled way to\nperform sampling with conformal guarantees.\n", "date": "2024-02-01", "categories": []} {"id": "2310.01262", "title": "Non-Exchangeable Conformal Risk Control", "abstract": " Split conformal prediction has recently sparked great interest due to its\nability to provide formally guaranteed uncertainty sets or intervals for\npredictions made by black-box neural models, ensuring a predefined probability\nof containing the actual ground truth. While the original formulation assumes\ndata exchangeability, some extensions handle non-exchangeable data, which is\noften the case in many real-world scenarios. In parallel, some progress has\nbeen made in conformal methods that provide statistical guarantees for a\nbroader range of objectives, such as bounding the best $F_1$-score or\nminimizing the false negative rate in expectation. In this paper, we leverage\nand extend these two lines of work by proposing non-exchangeable conformal risk\ncontrol, which allows controlling the expected value of any monotone loss\nfunction when the data is not exchangeable. Our framework is flexible, makes\nvery few assumptions, and allows weighting the data based on its relevance for\na given test example; a careful choice of weights may result on tighter bounds,\nmaking our framework useful in the presence of change points, time series, or\nother forms of distribution drift. Experiments with both synthetic and real\nworld data show the usefulness of our method.\n", "date": "2023-10-02", "categories": []} {"id": "2301.02424", "title": "Conformal Loss-Controlling Prediction", "abstract": " Conformal prediction is a learning framework controlling prediction coverage\nof prediction sets, which can be built on any learning algorithm for point\nprediction. This work proposes a learning framework named conformal\nloss-controlling prediction, which extends conformal prediction to the\nsituation where the value of a loss function needs to be controlled. Different\nfrom existing works about risk-controlling prediction sets and conformal risk\ncontrol with the purpose of controlling the expected values of loss functions,\nthe proposed approach in this paper focuses on the loss for any test object,\nwhich is an extension of conformal prediction from miscoverage loss to some\ngeneral loss. The controlling guarantee is proved under the assumption of\nexchangeability of data in finite-sample cases and the framework is tested\nempirically for classification with a class-varying loss and statistical\npostprocessing of numerical weather forecasting applications, which are\nintroduced as point-wise classification and point-wise regression problems. All\ntheoretical analysis and experimental results confirm the effectiveness of our\nloss-controlling approach.\n", "date": "2023-01-06", "categories": []} {"id": "2301.04378", "title": "Loss-Controlling Calibration for Predictive Models", "abstract": " We propose a learning framework for calibrating predictive models to make\nloss-controlling prediction for exchangeable data, which extends our recently\nproposed conformal loss-controlling prediction for more general cases. By\ncomparison, the predictors built by the proposed loss-controlling approach are\nnot limited to set predictors, and the loss function can be any measurable\nfunction without the monotone assumption. To control the loss values in an\nefficient way, we introduce transformations preserving exchangeability to prove\nfinite-sample controlling guarantee when the test label is obtained, and then\ndevelop an approximation approach to construct predictors. The transformations\ncan be built on any predefined function, which include using optimization\nalgorithms for parameter searching. This approach is a natural extension of\nconformal loss-controlling prediction, since it can be reduced to the latter\nwhen the set predictors have the nesting property and the loss functions are\nmonotone. Our proposed method is applied to selective regression and\nhigh-impact weather forecasting problems, which demonstrates its effectiveness\nfor general loss-controlling prediction.\n", "date": "2023-01-11", "categories": []} {"id": "2401.11810", "title": "Generalization and Informativeness of Conformal Prediction", "abstract": " The safe integration of machine learning modules in decision-making processes\nhinges on their ability to quantify uncertainty. A popular technique to achieve\nthis goal is conformal prediction (CP), which transforms an arbitrary base\npredictor into a set predictor with coverage guarantees. While CP certifies the\npredicted set to contain the target quantity with a user-defined tolerance, it\ndoes not provide control over the average size of the predicted sets, i.e.,\nover the informativeness of the prediction. In this work, a theoretical\nconnection is established between the generalization properties of the base\npredictor and the informativeness of the resulting CP prediction sets. To this\nend, an upper bound is derived on the expected size of the CP set predictor\nthat builds on generalization error bounds for the base predictor. The derived\nupper bound provides insights into the dependence of the average size of the CP\nset predictor on the amount of calibration data, the target reliability, and\nthe generalization performance of the base predictor. The theoretical insights\nare validated using simple numerical regression and classification tasks.\n", "date": "2024-01-22", "categories": []} {"id": "2401.07733", "title": "Conformal Approach To Gaussian Process Surrogate Evaluation With\n Coverage Guarantees", "abstract": " Gaussian processes (GPs) are a Bayesian machine learning approach widely used\nto construct surrogate models for the uncertainty quantification of computer\nsimulation codes in industrial applications. It provides both a mean predictor\nand an estimate of the posterior prediction variance, the latter being used to\nproduce Bayesian credibility intervals. Interpreting these intervals relies on\nthe Gaussianity of the simulation model as well as the well-specification of\nthe priors which are not always appropriate. We propose to address this issue\nwith the help of conformal prediction. In the present work, a method for\nbuilding adaptive cross-conformal prediction intervals is proposed by weighting\nthe non-conformity score with the posterior standard deviation of the GP. The\nresulting conformal prediction intervals exhibit a level of adaptivity akin to\nBayesian credibility sets and display a significant correlation with the\nsurrogate model local approximation error, while being free from the underlying\nmodel assumptions and having frequentist coverage guarantees. These estimators\ncan thus be used for evaluating the quality of a GP surrogate model and can\nassist a decision-maker in the choice of the best prior for the specific\napplication of the GP. The performance of the method is illustrated through a\npanel of numerical examples based on various reference databases. Moreover, the\npotential applicability of the method is demonstrated in the context of\nsurrogate modeling of an expensive-to-evaluate simulator of the clogging\nphenomenon in steam generators of nuclear reactors.\n", "date": "2024-01-15", "categories": []} {"id": "2401.06421", "title": "Uncertainty quantification for probabilistic machine learning in earth\n observation using conformal prediction", "abstract": " Unreliable predictions can occur when using artificial intelligence (AI)\nsystems with negative consequences for downstream applications, particularly\nwhen employed for decision-making. Conformal prediction provides a\nmodel-agnostic framework for uncertainty quantification that can be applied to\nany dataset, irrespective of its distribution, post hoc. In contrast to other\npixel-level uncertainty quantification methods, conformal prediction operates\nwithout requiring access to the underlying model and training dataset,\nconcurrently offering statistically valid and informative prediction regions,\nall while maintaining computational efficiency. In response to the increased\nneed to report uncertainty alongside point predictions, we bring attention to\nthe promise of conformal prediction within the domain of Earth Observation (EO)\napplications. To accomplish this, we assess the current state of uncertainty\nquantification in the EO domain and found that only 20% of the reviewed Google\nEarth Engine (GEE) datasets incorporated a degree of uncertainty information,\nwith unreliable methods prevalent. Next, we introduce modules that seamlessly\nintegrate into existing GEE predictive modelling workflows and demonstrate the\napplication of these tools for datasets spanning local to global scales,\nincluding the Dynamic World and Global Ecosystem Dynamics Investigation (GEDI)\ndatasets. These case studies encompass regression and classification tasks,\nfeaturing both traditional and deep learning-based workflows. Subsequently, we\ndiscuss the opportunities arising from the use of conformal prediction in EO.\nWe anticipate that the increased availability of easy-to-use implementations of\nconformal predictors, such as those provided here, will drive wider adoption of\nrigorous uncertainty quantification in EO, thereby enhancing the reliability of\nuses such as operational monitoring and decision making.\n", "date": "2024-01-12", "categories": []} {"id": "2304.01075", "title": "Conformal Prediction Regions for Time Series using Linear\n Complementarity Programming", "abstract": " Conformal prediction is a statistical tool for producing prediction regions\nof machine learning models that are valid with high probability. However,\napplying conformal prediction to time series data leads to conservative\nprediction regions. In fact, to obtain prediction regions over $T$ time steps\nwith confidence $1-\\delta$, {previous works require that each individual\nprediction region is valid} with confidence $1-\\delta/T$. We propose an\noptimization-based method for reducing this conservatism to enable long horizon\nplanning and verification when using learning-enabled time series predictors.\nInstead of considering prediction errors individually at each time step, we\nconsider a parameterized prediction error over multiple time steps. By\noptimizing the parameters over an additional dataset, we find prediction\nregions that are not conservative. We show that this problem can be cast as a\nmixed integer linear complementarity program (MILCP), which we then relax into\na linear complementarity program (LCP). Additionally, we prove that the relaxed\nLP has the same optimal cost as the original MILCP. Finally, we demonstrate the\nefficacy of our method on case studies using pedestrian trajectory predictors\nand F16 fighter jet altitude predictors.\n", "date": "2023-04-03", "categories": []} {"id": "2306.00876", "title": "Quantifying Deep Learning Model Uncertainty in Conformal Prediction", "abstract": " Precise estimation of predictive uncertainty in deep neural networks is a\ncritical requirement for reliable decision-making in machine learning and\nstatistical modeling, particularly in the context of medical AI. Conformal\nPrediction (CP) has emerged as a promising framework for representing the model\nuncertainty by providing well-calibrated confidence levels for individual\npredictions. However, the quantification of model uncertainty in conformal\nprediction remains an active research area, yet to be fully addressed. In this\npaper, we explore state-of-the-art CP methodologies and their theoretical\nfoundations. We propose a probabilistic approach in quantifying the model\nuncertainty derived from the produced prediction sets in conformal prediction\nand provide certified boundaries for the computed uncertainty. By doing so, we\nallow model uncertainty measured by CP to be compared by other uncertainty\nquantification methods such as Bayesian (e.g., MC-Dropout and DeepEnsemble) and\nEvidential approaches.\n", "date": "2023-06-01", "categories": []} {"id": "2302.03791", "title": "How to Trust Your Diffusion Model: A Convex Optimization Approach to\n Conformal Risk Control", "abstract": " Score-based generative modeling, informally referred to as diffusion models,\ncontinue to grow in popularity across several important domains and tasks.\nWhile they provide high-quality and diverse samples from empirical\ndistributions, important questions remain on the reliability and\ntrustworthiness of these sampling procedures for their responsible use in\ncritical scenarios. Conformal prediction is a modern tool to construct\nfinite-sample, distribution-free uncertainty guarantees for any black-box\npredictor. In this work, we focus on image-to-image regression tasks and we\npresent a generalization of the Risk-Controlling Prediction Sets (RCPS)\nprocedure, that we term $K$-RCPS, which allows to $(i)$ provide entrywise\ncalibrated intervals for future samples of any diffusion model, and $(ii)$\ncontrol a certain notion of risk with respect to a ground truth image with\nminimal mean interval length. Differently from existing conformal risk control\nprocedures, ours relies on a novel convex optimization approach that allows for\nmultidimensional risk control while provably minimizing the mean interval\nlength. We illustrate our approach on two real-world image denoising problems:\non natural images of faces as well as on computed tomography (CT) scans of the\nabdomen, demonstrating state of the art performance.\n", "date": "2023-02-07", "categories": []} {"id": "2312.09304", "title": "Well-calibrated Confidence Measures for Multi-label Text Classification\n with a Large Number of Labels", "abstract": " We extend our previous work on Inductive Conformal Prediction (ICP) for\nmulti-label text classification and present a novel approach for addressing the\ncomputational inefficiency of the Label Powerset (LP) ICP, arrising when\ndealing with a high number of unique labels. We present experimental results\nusing the original and the proposed efficient LP-ICP on two English and one\nCzech language data-sets. Specifically, we apply the LP-ICP on three deep\nArtificial Neural Network (ANN) classifiers of two types: one based on\ncontextualised (bert) and two on non-contextualised (word2vec) word-embeddings.\nIn the LP-ICP setting we assign nonconformity scores to label-sets from which\nthe corresponding p-values and prediction-sets are determined. Our approach\ndeals with the increased computational burden of LP by eliminating from\nconsideration a significant number of label-sets that will surely have p-values\nbelow the specified significance level. This reduces dramatically the\ncomputational complexity of the approach while fully respecting the standard CP\nguarantees. Our experimental results show that the contextualised-based\nclassifier surpasses the non-contextualised-based ones and obtains\nstate-of-the-art performance for all data-sets examined. The good performance\nof the underlying classifiers is carried on to their ICP counterparts without\nany significant accuracy loss, but with the added benefits of ICP, i.e. the\nconfidence information encapsulated in the prediction sets. We experimentally\ndemonstrate that the resulting prediction sets can be tight enough to be\npractically useful even though the set of all possible label-sets contains more\nthan $1e+16$ combinations. Additionally, the empirical error rates of the\nobtained prediction-sets confirm that our outputs are well-calibrated.\n", "date": "2023-12-14", "categories": []} {"id": "2312.09606", "title": "Reliable Prediction Intervals with Regression Neural Networks", "abstract": " This paper proposes an extension to conventional regression Neural Networks\n(NNs) for replacing the point predictions they produce with prediction\nintervals that satisfy a required level of confidence. Our approach follows a\nnovel machine learning framework, called Conformal Prediction (CP), for\nassigning reliable confidence measures to predictions without assuming anything\nmore than that the data are independent and identically distributed (i.i.d.).\nWe evaluate the proposed method on four benchmark datasets and on the problem\nof predicting Total Electron Content (TEC), which is an important parameter in\ntrans-ionospheric links; for the latter we use a dataset of more than 60000 TEC\nmeasurements collected over a period of 11 years. Our experimental results show\nthat the prediction intervals produced by our method are both well-calibrated\nand tight enough to be useful in practice.\n", "date": "2023-12-15", "categories": []} {"id": "2210.12496", "title": "Bayesian Optimization with Conformal Prediction Sets", "abstract": " Bayesian optimization is a coherent, ubiquitous approach to decision-making\nunder uncertainty, with applications including multi-arm bandits, active\nlearning, and black-box optimization. Bayesian optimization selects decisions\n(i.e. objective function queries) with maximal expected utility with respect to\nthe posterior distribution of a Bayesian model, which quantifies reducible,\nepistemic uncertainty about query outcomes. In practice, subjectively\nimplausible outcomes can occur regularly for two reasons: 1) model\nmisspecification and 2) covariate shift. Conformal prediction is an uncertainty\nquantification method with coverage guarantees even for misspecified models and\na simple mechanism to correct for covariate shift. We propose conformal\nBayesian optimization, which directs queries towards regions of search space\nwhere the model predictions have guaranteed validity, and investigate its\nbehavior on a suite of black-box optimization tasks and tabular ranking tasks.\nIn many cases we find that query coverage can be significantly improved without\nharming sample-efficiency.\n", "date": "2022-10-22", "categories": []} {"id": "2312.04658", "title": "PAC-Bayes Generalization Certificates for Learned Inductive Conformal\n Prediction", "abstract": " Inductive Conformal Prediction (ICP) provides a practical and effective\napproach for equipping deep learning models with uncertainty estimates in the\nform of set-valued predictions which are guaranteed to contain the ground truth\nwith high probability. Despite the appeal of this coverage guarantee, these\nsets may not be efficient: the size and contents of the prediction sets are not\ndirectly controlled, and instead depend on the underlying model and choice of\nscore function. To remedy this, recent work has proposed learning model and\nscore function parameters using data to directly optimize the efficiency of the\nICP prediction sets. While appealing, the generalization theory for such an\napproach is lacking: direct optimization of empirical efficiency may yield\nprediction sets that are either no longer efficient on test data, or no longer\nobtain the required coverage on test data. In this work, we use PAC-Bayes\ntheory to obtain generalization bounds on both the coverage and the efficiency\nof set-valued predictors which can be directly optimized to maximize efficiency\nwhile satisfying a desired test coverage. In contrast to prior work, our\nframework allows us to utilize the entire calibration dataset to learn the\nparameters of the model and score function, instead of requiring a separate\nhold-out set for obtaining test-time coverage guarantees. We leverage these\ntheoretical results to provide a practical algorithm for using calibration data\nto simultaneously fine-tune the parameters of a model and score function while\nguaranteeing test-time coverage and efficiency of the resulting prediction\nsets. We evaluate the approach on regression and classification tasks, and\noutperform baselines calibrated using a Hoeffding bound-based PAC guarantee on\nICP, especially in the low-data regime.\n", "date": "2023-12-07", "categories": []} {"id": "2312.04950", "title": "Sequential inductive prediction intervals", "abstract": " In this paper we explore the concept of sequential inductive prediction\nintervals using theory from sequential testing. We furthermore introduce a\n3-parameter PAC definition of prediction intervals that allows us via\nsimulation to achieve almost sharp bounds with high probability.\n", "date": "2023-12-08", "categories": []} {"id": "2312.05195", "title": "Conformal Prediction in Multi-User Settings: An Evaluation", "abstract": " Typically, machine learning models are trained and evaluated without making\nany distinction between users (e.g, using traditional hold-out and\ncross-validation). However, this produces inaccurate performance metrics\nestimates in multi-user settings. That is, situations where the data were\ncollected by multiple users with different characteristics (e.g., age, gender,\nheight, etc.) which is very common in user computer interaction and medical\napplications. For these types of scenarios model evaluation strategies that\nprovide better performance estimates have been proposed such as mixed,\nuser-independent, user-dependent, and user-adaptive models. Although those\nstrategies are better suited for multi-user systems, they are typically\nassessed with respect to performance metrics that capture the overall behavior\nof the models and do not provide any performance guarantees for individual\npredictions nor they provide any feedback about the predictions' uncertainty.\nIn order to overcome those limitations, in this work we evaluated the conformal\nprediction framework in several multi-user settings. Conformal prediction is a\nmodel agnostic method that provides confidence guarantees on the predictions,\nthus, increasing the trustworthiness and robustness of the models. We conducted\nextensive experiments using different evaluation strategies and found\nsignificant differences in terms of conformal performance measures. We also\nproposed several visualizations based on matrices, graphs, and charts that\ncapture different aspects of the resulting prediction sets.\n", "date": "2023-12-08", "categories": []} {"id": "2310.11479", "title": "On the Temperature of Bayesian Graph Neural Networks for Conformal\n Prediction", "abstract": " Accurate uncertainty quantification in graph neural networks (GNNs) is\nessential, especially in high-stakes domains where GNNs are frequently\nemployed. Conformal prediction (CP) offers a promising framework for\nquantifying uncertainty by providing $\\textit{valid}$ prediction sets for any\nblack-box model. CP ensures formal probabilistic guarantees that a prediction\nset contains a true label with a desired probability. However, the size of\nprediction sets, known as $\\textit{inefficiency}$, is influenced by the\nunderlying model and data generating process. On the other hand, Bayesian\nlearning also provides a credible region based on the estimated posterior\ndistribution, but this region is $\\textit{well-calibrated}$ only when the model\nis correctly specified. Building on a recent work that introduced a scaling\nparameter for constructing valid credible regions from posterior estimate, our\nstudy explores the advantages of incorporating a temperature parameter into\nBayesian GNNs within CP framework. We empirically demonstrate the existence of\ntemperatures that result in more efficient prediction sets. Furthermore, we\nconduct an analysis to identify the factors contributing to inefficiency and\noffer valuable insights into the relationship between CP performance and model\ncalibration.\n", "date": "2023-10-17", "categories": []} {"id": "2312.01959", "title": "Learning-Based Approaches to Predictive Monitoring with Conformal\n Statistical Guarantees", "abstract": " This tutorial focuses on efficient methods to predictive monitoring (PM), the\nproblem of detecting at runtime future violations of a given requirement from\nthe current state of a system. While performing model checking at runtime would\noffer a precise solution to the PM problem, it is generally computationally\nexpensive. To address this scalability issue, several lightweight approaches\nbased on machine learning have recently been proposed. These approaches work by\nlearning an approximate yet efficient surrogate (deep learning) model of the\nexpensive model checker. A key challenge remains to ensure reliable\npredictions, especially in safety-critical applications. We review our recent\nwork on predictive monitoring, one of the first to propose learning-based\napproximations for CPS verification of temporal logic specifications and the\nfirst in this context to apply conformal prediction (CP) for rigorous\nuncertainty quantification. These CP-based uncertainty estimators offer\nstatistical guarantees regarding the generalization error of the learning\nmodel, and they can be used to determine unreliable predictions that should be\nrejected. In this tutorial, we present a general and comprehensive framework\nsummarizing our approach to the predictive monitoring of CPSs, examining in\ndetail several variants determined by three main dimensions: system dynamics\n(deterministic, non-deterministic, stochastic), state observability, and\nsemantics of requirements' satisfaction (Boolean or quantitative).\n", "date": "2023-12-04", "categories": []} {"id": "2311.11837", "title": "Kandinsky Conformal Prediction: Efficient Calibration of Image\n Segmentation Algorithms", "abstract": " Image segmentation algorithms can be understood as a collection of pixel\nclassifiers, for which the outcomes of nearby pixels are correlated. Classifier\nmodels can be calibrated using Inductive Conformal Prediction, but this\nrequires holding back a sufficiently large calibration dataset for computing\nthe distribution of non-conformity scores of the model's predictions. If one\nonly requires only marginal calibration on the image level, this calibration\nset consists of all individual pixels in the images available for calibration.\nHowever, if the goal is to attain proper calibration for each individual pixel\nclassifier, the calibration set consists of individual images. In a scenario\nwhere data are scarce (such as the medical domain), it may not always be\npossible to set aside sufficiently many images for this pixel-level\ncalibration. The method we propose, dubbed ``Kandinsky calibration'', makes use\nof the spatial structure present in the distribution of natural images to\nsimultaneously calibrate the classifiers of ``similar'' pixels. This can be\nseen as an intermediate approach between marginal (imagewise) and conditional\n(pixelwise) calibration, where non-conformity scores are aggregated over\nsimilar image regions, thereby making more efficient use of the images\navailable for calibration. We run experiments on segmentation algorithms\ntrained and calibrated on subsets of the public MS-COCO and Medical Decathlon\ndatasets, demonstrating that Kandinsky calibration method can significantly\nimprove the coverage. When compared to both pixelwise and imagewise calibration\non little data, the Kandinsky method achieves much lower coverage errors,\nindicating the data efficiency of the Kandinsky calibration.\n", "date": "2023-11-20", "categories": []} {"id": "2303.12783", "title": "Conformal Prediction for Time Series with Modern Hopfield Networks", "abstract": " To quantify uncertainty, conformal prediction methods are gaining\ncontinuously more interest and have already been successfully applied to\nvarious domains. However, they are difficult to apply to time series as the\nautocorrelative structure of time series violates basic assumptions required by\nconformal prediction. We propose HopCPT, a novel conformal prediction approach\nfor time series that not only copes with temporal structures but leverages\nthem. We show that our approach is theoretically well justified for time series\nwhere temporal dependencies are present. In experiments, we demonstrate that\nour new approach outperforms state-of-the-art conformal prediction methods on\nmultiple real-world time series datasets from four different domains.\n", "date": "2023-03-22", "categories": []} {"id": "2311.00774", "title": "Conformalized Deep Splines for Optimal and Efficient Prediction Sets", "abstract": " Uncertainty estimation is critical in high-stakes machine learning\napplications. One effective way to estimate uncertainty is conformal\nprediction, which can provide predictive inference with statistical coverage\nguarantees. We present a new conformal regression method, Spline Prediction\nIntervals via Conformal Estimation (SPICE), that estimates the conditional\ndensity using neural-network-parameterized splines. We prove universal\napproximation and optimality results for SPICE, which are empirically validated\nby our experiments. SPICE is compatible with two different efficient-to-compute\nconformal scores, one oracle-optimal for marginal coverage (SPICE-ND) and the\nother asymptotically optimal for conditional coverage (SPICE-HPD). Results on\nbenchmark datasets demonstrate SPICE-ND models achieve the smallest average\nprediction set sizes, including average size reductions of nearly 50% for some\ndatasets compared to the next best baseline. SPICE-HPD models achieve the best\nconditional coverage compared to baselines. The SPICE implementation is made\navailable.\n", "date": "2023-11-01", "categories": []} {"id": "2207.02808", "title": "Improved conformalized quantile regression", "abstract": " Conformalized quantile regression is a procedure that inherits the advantages\nof conformal prediction and quantile regression. That is, we use quantile\nregression to estimate the true conditional quantile and then apply a conformal\nstep on a calibration set to ensure marginal coverage. In this way, we get\nadaptive prediction intervals that account for heteroscedasticity. However, the\naforementioned conformal step lacks adaptiveness as described in (Romano et\nal., 2019). To overcome this limitation, instead of applying a single conformal\nstep after estimating conditional quantiles with quantile regression, we\npropose to cluster the explanatory variables weighted by their permutation\nimportance with an optimized k-means and apply k conformal steps. To show that\nthis improved version outperforms the classic version of conformalized quantile\nregression and is more adaptive to heteroscedasticity, we extensively compare\nthe prediction intervals of both in open datasets.\n", "date": "2022-07-06", "categories": []} {"id": "2306.09335", "title": "Class-Conditional Conformal Prediction with Many Classes", "abstract": " Standard conformal prediction methods provide a marginal coverage guarantee,\nwhich means that for a random test point, the conformal prediction set contains\nthe true label with a user-specified probability. In many classification\nproblems, we would like to obtain a stronger guarantee--that for test points of\na specific class, the prediction set contains the true label with the same\nuser-chosen probability. For the latter goal, existing conformal prediction\nmethods do not work well when there is a limited amount of labeled data per\nclass, as is often the case in real applications where the number of classes is\nlarge. We propose a method called clustered conformal prediction that clusters\ntogether classes having \"similar\" conformal scores and performs conformal\nprediction at the cluster level. Based on empirical evaluation across four\nimage data sets with many (up to 1000) classes, we find that clustered\nconformal typically outperforms existing methods in terms of class-conditional\ncoverage and set size metrics.\n", "date": "2023-06-15", "categories": []} {"id": "2302.07294", "title": "Derandomized Novelty Detection with FDR Control via Conformal E-values", "abstract": " Conformal inference provides a general distribution-free method to rigorously\ncalibrate the output of any machine learning algorithm for novelty detection.\nWhile this approach has many strengths, it has the limitation of being\nrandomized, in the sense that it may lead to different results when analyzing\ntwice the same data, and this can hinder the interpretation of any findings. We\npropose to make conformal inferences more stable by leveraging suitable\nconformal e-values instead of p-values to quantify statistical significance.\nThis solution allows the evidence gathered from multiple analyses of the same\ndata to be aggregated effectively while provably controlling the false\ndiscovery rate. Further, we show that the proposed method can reduce randomness\nwithout much loss of power compared to standard conformal inference, partly\nthanks to an innovative way of weighting conformal e-values based on additional\nside information carefully extracted from the same data. Simulations with\nsynthetic and real data confirm this solution can be effective at eliminating\nrandom noise in the inferences obtained with state-of-the-art alternative\ntechniques, sometimes also leading to higher power.\n", "date": "2023-02-14", "categories": []} {"id": "2307.09302", "title": "Conformal prediction under ambiguous ground truth", "abstract": " Conformal Prediction (CP) allows to perform rigorous uncertainty\nquantification by constructing a prediction set $C(X)$ satisfying $\\mathbb{P}(Y\n\\in C(X))\\geq 1-\\alpha$ for a user-chosen $\\alpha \\in [0,1]$ by relying on\ncalibration data $(X_1,Y_1),...,(X_n,Y_n)$ from $\\mathbb{P}=\\mathbb{P}^{X}\n\\otimes \\mathbb{P}^{Y|X}$. It is typically implicitly assumed that\n$\\mathbb{P}^{Y|X}$ is the \"true\" posterior label distribution. However, in many\nreal-world scenarios, the labels $Y_1,...,Y_n$ are obtained by aggregating\nexpert opinions using a voting procedure, resulting in a one-hot distribution\n$\\mathbb{P}_{vote}^{Y|X}$. For such ``voted'' labels, CP guarantees are thus\nw.r.t. $\\mathbb{P}_{vote}=\\mathbb{P}^X \\otimes \\mathbb{P}_{vote}^{Y|X}$ rather\nthan the true distribution $\\mathbb{P}$. In cases with unambiguous ground truth\nlabels, the distinction between $\\mathbb{P}_{vote}$ and $\\mathbb{P}$ is\nirrelevant. However, when experts do not agree because of ambiguous labels,\napproximating $\\mathbb{P}^{Y|X}$ with a one-hot distribution\n$\\mathbb{P}_{vote}^{Y|X}$ ignores this uncertainty. In this paper, we propose\nto leverage expert opinions to approximate $\\mathbb{P}^{Y|X}$ using a\nnon-degenerate distribution $\\mathbb{P}_{agg}^{Y|X}$. We develop Monte Carlo CP\nprocedures which provide guarantees w.r.t. $\\mathbb{P}_{agg}=\\mathbb{P}^X\n\\otimes \\mathbb{P}_{agg}^{Y|X}$ by sampling multiple synthetic pseudo-labels\nfrom $\\mathbb{P}_{agg}^{Y|X}$ for each calibration example $X_1,...,X_n$. In a\ncase study of skin condition classification with significant disagreement among\nexpert annotators, we show that applying CP w.r.t. $\\mathbb{P}_{vote}$\nunder-covers expert annotations: calibrated for $72\\%$ coverage, it falls short\nby on average $10\\%$; our Monte Carlo CP closes this gap both empirically and\ntheoretically.\n", "date": "2023-07-18", "categories": []} {"id": "2304.03398", "title": "Quantum Conformal Prediction for Reliable Uncertainty Quantification in\n Quantum Machine Learning", "abstract": " In this work, we aim at augmenting the decisions output by quantum models\nwith \"error bars\" that provide finite-sample coverage guarantees. Quantum\nmodels implement implicit probabilistic predictors that produce multiple random\ndecisions for each input through measurement shots. Randomness arises not only\nfrom the inherent stochasticity of quantum measurements, but also from quantum\ngate noise and quantum measurement noise caused by noisy hardware. Furthermore,\nquantum noise may be correlated across shots and it may present drifts in time.\nThis paper proposes to leverage such randomness to define prediction sets for\nboth classification and regression that provably capture the uncertainty of the\nmodel. The approach builds on probabilistic conformal prediction (PCP), while\naccounting for the unique features of quantum models. Among the key technical\ninnovations, we introduce a new general class of non-conformity scores that\naddress the presence of quantum noise, including possible drifts. Experimental\nresults, using both simulators and current quantum computers, confirm the\ntheoretical calibration guarantees of the proposed framework.\n", "date": "2023-04-06", "categories": []} {"id": "2310.12964", "title": "PAC Prediction Sets Under Label Shift", "abstract": " Prediction sets capture uncertainty by predicting sets of labels rather than\nindividual labels, enabling downstream decisions to conservatively account for\nall plausible outcomes. Conformal inference algorithms construct prediction\nsets guaranteed to contain the true label with high probability. These\nguarantees fail to hold in the face of distribution shift, which is precisely\nwhen reliable uncertainty quantification can be most useful. We propose a novel\nalgorithm for constructing prediction sets with PAC guarantees in the label\nshift setting. This method estimates the predicted probabilities of the classes\nin a target domain, as well as the confusion matrix, then propagates\nuncertainty in these estimates through a Gaussian elimination algorithm to\ncompute confidence intervals for importance weights. Finally, it uses these\nintervals to construct prediction sets. We evaluate our approach on five\ndatasets: the CIFAR-10, ChestX-Ray and Entity-13 image datasets, the tabular\nCDC Heart dataset, and the AGNews text dataset. Our algorithm satisfies the PAC\nguarantee while producing smaller, more informative, prediction sets compared\nto several baselines.\n", "date": "2023-10-19", "categories": []} {"id": "2310.12033", "title": "Conformal Drug Property Prediction with Density Estimation under\n Covariate Shift", "abstract": " In drug discovery, it is vital to confirm the predictions of pharmaceutical\nproperties from computational models using costly wet-lab experiments. Hence,\nobtaining reliable uncertainty estimates is crucial for prioritizing drug\nmolecules for subsequent experimental validation. Conformal Prediction (CP) is\na promising tool for creating such prediction sets for molecular properties\nwith a coverage guarantee. However, the exchangeability assumption of CP is\noften challenged with covariate shift in drug discovery tasks: Most datasets\ncontain limited labeled data, which may not be representative of the vast\nchemical space from which molecules are drawn. To address this limitation, we\npropose a method called CoDrug that employs an energy-based model leveraging\nboth training data and unlabelled data, and Kernel Density Estimation (KDE) to\nassess the densities of a molecule set. The estimated densities are then used\nto weigh the molecule samples while building prediction sets and rectifying for\ndistribution shift. In extensive experiments involving realistic distribution\ndrifts in various small-molecule drug discovery tasks, we demonstrate the\nability of CoDrug to provide valid prediction sets and its utility in\naddressing the distribution shift arising from de novo drug design models. On\naverage, using CoDrug can reduce the coverage gap by over 35% when compared to\nconformal prediction sets not adjusted for covariate shift.\n", "date": "2023-10-18", "categories": []} {"id": "2310.10299", "title": "Forking Uncertainties: Reliable Prediction and Model Predictive Control\n with Sequence Models via Conformal Risk Control", "abstract": " In many real-world problems, predictions are leveraged to monitor and control\ncyber-physical systems, demanding guarantees on the satisfaction of reliability\nand safety requirements. However, predictions are inherently uncertain, and\nmanaging prediction uncertainty presents significant challenges in environments\ncharacterized by complex dynamics and forking trajectories. In this work, we\nassume access to a pre-designed probabilistic implicit or explicit sequence\nmodel, which may have been obtained using model-based or model-free methods. We\nintroduce probabilistic time series-conformal risk prediction (PTS-CRC), a\nnovel post-hoc calibration procedure that operates on the predictions produced\nby any pre-designed probabilistic forecaster to yield reliable error bars. In\ncontrast to existing art, PTS-CRC produces predictive sets based on an ensemble\nof multiple prototype trajectories sampled from the sequence model, supporting\nthe efficient representation of forking uncertainties. Furthermore, unlike the\nstate of the art, PTS-CRC can satisfy reliability definitions beyond coverage.\nThis property is leveraged to devise a novel model predictive control (MPC)\nframework that addresses open-loop and closed-loop control problems under\ngeneral average constraints on the quality or safety of the control policy. We\nexperimentally validate the performance of PTS-CRC prediction and control by\nstudying a number of use cases in the context of wireless networking. Across\nall the considered tasks, PTS-CRC predictors are shown to provide more\ninformative predictive sets, as well as safe control policies with larger\nreturns.\n", "date": "2023-10-16", "categories": []} {"id": "2208.08401", "title": "Conformal Inference for Online Prediction with Arbitrary Distribution\n Shifts", "abstract": " We consider the problem of forming prediction sets in an online setting where\nthe distribution generating the data is allowed to vary over time. Previous\napproaches to this problem suffer from over-weighting historical data and thus\nmay fail to quickly react to the underlying dynamics. Here we correct this\nissue and develop a novel procedure with provably small regret over all local\ntime intervals of a given width. We achieve this by modifying the adaptive\nconformal inference (ACI) algorithm of Gibbs and Cand\\`{e}s (2021) to contain\nan additional step in which the step-size parameter of ACI's gradient descent\nupdate is tuned over time. Crucially, this means that unlike ACI, which\nrequires knowledge of the rate of change of the data-generating mechanism, our\nnew procedure is adaptive to both the size and type of the distribution shift.\nOur methods are highly flexible and can be used in combination with any\nbaseline predictive algorithm that produces point estimates or estimated\nquantiles of the target without the need for distributional assumptions. We\ntest our techniques on two real-world datasets aimed at predicting stock market\nvolatility and COVID-19 case counts and find that they are robust and adaptive\nto real-world distribution shifts.\n", "date": "2022-08-17", "categories": []} {"id": "2310.03545", "title": "Distribution-free risk assessment of regression-based machine learning\n algorithms", "abstract": " Machine learning algorithms have grown in sophistication over the years and\nare increasingly deployed for real-life applications. However, when using\nmachine learning techniques in practical settings, particularly in high-risk\napplications such as medicine and engineering, obtaining the failure\nprobability of the predictive model is critical. We refer to this problem as\nthe risk-assessment task. We focus on regression algorithms and the\nrisk-assessment task of computing the probability of the true label lying\ninside an interval defined around the model's prediction. We solve the\nrisk-assessment problem using the conformal prediction approach, which provides\nprediction intervals that are guaranteed to contain the true label with a given\nprobability. Using this coverage property, we prove that our approximated\nfailure probability is conservative in the sense that it is not lower than the\ntrue failure probability of the ML algorithm. We conduct extensive experiments\nto empirically study the accuracy of the proposed method for problems with and\nwithout covariate shift. Our analysis focuses on different modeling regimes,\ndataset sizes, and conformal prediction methodologies.\n", "date": "2023-10-05", "categories": []} {"id": "2310.02863", "title": "Conformal Predictions for Longitudinal Data", "abstract": " We introduce Longitudinal Predictive Conformal Inference (LPCI), a novel\ndistribution-free conformal prediction algorithm for longitudinal data. Current\nconformal prediction approaches for time series data predominantly focus on the\nunivariate setting, and thus lack cross-sectional coverage when applied\nindividually to each time series in a longitudinal dataset. The current\nstate-of-the-art for longitudinal data relies on creating infinitely-wide\nprediction intervals to guarantee both cross-sectional and asymptotic\nlongitudinal coverage. The proposed LPCI method addresses this by ensuring that\nboth longitudinal and cross-sectional coverages are guaranteed without\nresorting to infinitely wide intervals. In our approach, we model the residual\ndata as a quantile fixed-effects regression problem, constructing prediction\nintervals with a trained quantile regressor. Our extensive experiments\ndemonstrate that LPCI achieves valid cross-sectional coverage and outperforms\nexisting benchmarks in terms of longitudinal coverage rates. Theoretically, we\nestablish LPCI's asymptotic coverage guarantees for both dimensions, with\nfinite-width intervals. The robust performance of LPCI in generating reliable\nprediction intervals for longitudinal data underscores its potential for broad\napplications, including in medicine, finance, and supply chain management.\n", "date": "2023-10-04", "categories": []} {"id": "2309.12510", "title": "Confidence Calibration for Systems with Cascaded Predictive Modules", "abstract": " Existing conformal prediction algorithms estimate prediction intervals at\ntarget confidence levels to characterize the performance of a regression model\non new test samples. However, considering an autonomous system consisting of\nmultiple modules, prediction intervals constructed for individual modules fall\nshort of accommodating uncertainty propagation over different modules and thus\ncannot provide reliable predictions on system behavior. We address this\nlimitation and present novel solutions based on conformal prediction to provide\nprediction intervals calibrated for a predictive system consisting of cascaded\nmodules (e.g., an upstream feature extraction module and a downstream\nregression module). Our key idea is to leverage module-level validation data to\ncharacterize the system-level error distribution without direct access to\nend-to-end validation data. We provide theoretical justification and empirical\nexperimental results to demonstrate the effectiveness of proposed solutions. In\ncomparison to prediction intervals calibrated for individual modules, our\nsolutions generate improved intervals with more accurate performance guarantees\nfor system predictions, which are demonstrated on both synthetic systems and\nreal-world systems performing overlap prediction for indoor navigation using\nthe Matterport3D dataset.\n", "date": "2023-09-21", "categories": []} {"id": "2309.11018", "title": "Conformalized Multimodal Uncertainty Regression and Reasoning", "abstract": " This paper introduces a lightweight uncertainty estimator capable of\npredicting multimodal (disjoint) uncertainty bounds by integrating conformal\nprediction with a deep-learning regressor. We specifically discuss its\napplication for visual odometry (VO), where environmental features such as\nflying domain symmetries and sensor measurements under ambiguities and\nocclusion can result in multimodal uncertainties. Our simulation results show\nthat uncertainty estimates in our framework adapt sample-wise against\nchallenging operating conditions such as pronounced noise, limited training\ndata, and limited parametric size of the prediction model. We also develop a\nreasoning framework that leverages these robust uncertainty estimates and\nincorporates optical flow-based reasoning to improve prediction prediction\naccuracy. Thus, by appropriately accounting for predictive uncertainties of\ndata-driven learning and closing their estimation loop via rule-based\nreasoning, our methodology consistently surpasses conventional deep learning\napproaches on all these challenging scenarios--pronounced noise, limited\ntraining data, and limited model size-reducing the prediction error by 2-3x.\n", "date": "2023-09-20", "categories": []} {"id": "2304.03778", "title": "Conformal Regression in Calorie Prediction for Team Jumbo-Visma", "abstract": " UCI WorldTour races, the premier men's elite road cycling tour, are grueling\nevents that put physical fitness and endurance of riders to the test. The\ncoaches of Team Jumbo-Visma have long been responsible for predicting the\nenergy needs of each rider of the Dutch team for every race on the calendar.\nThose must be estimated to ensure riders have the energy and resources\nnecessary to maintain a high level of performance throughout a race. This task,\nhowever, is both time-consuming and challenging, as it requires precise\nestimates of race speed and power output. Traditionally, the approach to\npredicting energy needs has relied on judgement and experience of coaches, but\nthis method has its limitations and often leads to inaccurate predictions. In\nthis paper, we propose a new, more effective approach to predicting energy\nneeds for cycling races. By predicting the speed and power with regression\nmodels, we provide the coaches with calorie needs estimates for each individual\nrider per stage instantly. In addition, we compare methods to quantify\nuncertainty using conformal prediction. The empirical analysis of the\njackknife+, jackknife-minmax, jackknife-minmax-after-bootstrap, CV+, CV-minmax,\nconformalized quantile regression, and inductive conformal prediction methods\nin conformal prediction reveals that all methods achieve valid prediction\nintervals. All but minmax-based methods also produce sufficiently narrow\nprediction intervals for decision-making. Furthermore, methods computing\nprediction intervals of fixed size produce tighter intervals for low\nsignificance values. Among the methods computing intervals of varying length\nacross the input space, inductive conformal prediction computes narrower\nprediction intervals at larger significance level.\n", "date": "2023-04-06", "categories": []} {"id": "2309.04760", "title": "RR-CP: Reliable-Region-Based Conformal Prediction for Trustworthy\n Medical Image Classification", "abstract": " Conformal prediction (CP) generates a set of predictions for a given test\nsample such that the prediction set almost always contains the true label\n(e.g., 99.5\\% of the time). CP provides comprehensive predictions on possible\nlabels of a given test sample, and the size of the set indicates how certain\nthe predictions are (e.g., a set larger than one is `uncertain'). Such distinct\nproperties of CP enable effective collaborations between human experts and\nmedical AI models, allowing efficient intervention and quality check in\nclinical decision-making. In this paper, we propose a new method called\nReliable-Region-Based Conformal Prediction (RR-CP), which aims to impose a\nstronger statistical guarantee so that the user-specified error rate (e.g.,\n0.5\\%) can be achieved in the test time, and under this constraint, the size of\nthe prediction set is optimized (to be small). We consider a small prediction\nset size an important measure only when the user-specified error rate is\nachieved. Experiments on five public datasets show that our RR-CP performs\nwell: with a reasonably small-sized prediction set, it achieves the\nuser-specified error rate (e.g., 0.5\\%) significantly more frequently than\nexiting CP methods.\n", "date": "2023-09-09", "categories": []} {"id": "2309.03797", "title": "Conformal Autoregressive Generation: Beam Search with Coverage\n Guarantees", "abstract": " We introduce two new extensions to the beam search algorithm based on\nconformal predictions (CP) to produce sets of sequences with theoretical\ncoverage guarantees. The first method is very simple and proposes\ndynamically-sized subsets of beam search results but, unlike typical CP\nprocedures, has an upper bound on the achievable guarantee depending on a\npost-hoc calibration measure. Our second algorithm introduces the conformal set\nprediction procedure as part of the decoding process, producing a variable beam\nwidth which adapts to the current uncertainty. While more complex, this\nprocedure can achieve coverage guarantees selected a priori. We provide\nmarginal coverage bounds for each method, and evaluate them empirically on a\nselection of tasks drawing from natural language processing and chemistry.\n", "date": "2023-09-07", "categories": []} {"id": "2307.04993", "title": "Uncertainty Quantification of the Virial Black Hole Mass with Conformal\n Prediction", "abstract": " Precise measurements of the black hole mass are essential to gain insight on\nthe black hole and host galaxy co-evolution. A direct measure of the black hole\nmass is often restricted to nearest galaxies and instead, an indirect method\nusing the single-epoch virial black hole mass estimation is used for objects at\nhigh redshifts. However, this method is subjected to biases and uncertainties\nas it is reliant on the scaling relation from a small sample of local active\ngalactic nuclei. In this study, we propose the application of conformalised\nquantile regression (CQR) to quantify the uncertainties of the black hole\npredictions in a machine learning setting. We compare CQR with various\nprediction interval techniques and demonstrated that CQR can provide a more\nuseful prediction interval indicator. In contrast to baseline approaches for\nprediction interval estimation, we show that the CQR method provides prediction\nintervals that adjust to the black hole mass and its related properties. That\nis it yields a tighter constraint on the prediction interval (hence more\ncertain) for a larger black hole mass, and accordingly, bright and broad\nspectral line width source. Using a combination of neural network model and CQR\nframework, the recovered virial black hole mass predictions and uncertainties\nare comparable to those measured from the Sloan Digital Sky Survey. The code is\npublicly available at https://github.com/yongsukyee/uncertain_blackholemass.\n", "date": "2023-07-11", "categories": []} {"id": "2308.15094", "title": "Group-Conditional Conformal Prediction via Quantile Regression\n Calibration for Crop and Weed Classification", "abstract": " As deep learning predictive models become an integral part of a large\nspectrum of precision agricultural systems, a barrier to the adoption of such\nautomated solutions is the lack of user trust in these highly complex, opaque\nand uncertain models. Indeed, deep neural networks are not equipped with any\nexplicit guarantees that can be used to certify the system's performance,\nespecially in highly varying uncontrolled environments such as the ones\ntypically faced in computer vision for agriculture.Fortunately, certain methods\ndeveloped in other communities can prove to be important for agricultural\napplications. This article presents the conformal prediction framework that\nprovides valid statistical guarantees on the predictive performance of any\nblack box prediction machine, with almost no assumptions, applied to the\nproblem of deep visual classification of weeds and crops in real-world\nconditions. The framework is exposed with a focus on its practical aspects and\nspecial attention accorded to the Adaptive Prediction Sets (APS) approach that\ndelivers marginal guarantees on the model's coverage. Marginal results are then\nshown to be insufficient to guarantee performance on all groups of individuals\nin the population as characterized by their environmental and pedo-climatic\nauxiliary data gathered during image acquisition.To tackle this shortcoming,\ngroup-conditional conformal approaches are presented: the ''classical'' method\nthat consists of iteratively applying the APS procedure on all groups, and a\nproposed elegant reformulation and implementation of the procedure using\nquantile regression on group membership indicators. Empirical results showing\nthe validity of the proposed approach are presented and compared to the\nmarginal APS then discussed.\n", "date": "2023-08-29", "categories": []} {"id": "2304.06052", "title": "Confident Object Detection via Conformal Prediction and Conformal Risk\n Control: an Application to Railway Signaling", "abstract": " Deploying deep learning models in real-world certified systems requires the\nability to provide confidence estimates that accurately reflect their\nuncertainty. In this paper, we demonstrate the use of the conformal prediction\nframework to construct reliable and trustworthy predictors for detecting\nrailway signals. Our approach is based on a novel dataset that includes images\ntaken from the perspective of a train operator and state-of-the-art object\ndetectors. We test several conformal approaches and introduce a new method\nbased on conformal risk control. Our findings demonstrate the potential of the\nconformal prediction framework to evaluate model performance and provide\npractical guidance for achieving formally guaranteed uncertainty bounds.\n", "date": "2023-04-12", "categories": []} {"id": "2207.12377", "title": "A novel Deep Learning approach for one-step Conformal Prediction\n approximation", "abstract": " Deep Learning predictions with measurable confidence are increasingly\ndesirable for real-world problems, especially in high-risk settings. The\nConformal Prediction (CP) framework is a versatile solution that guarantees a\nmaximum error rate given minimal constraints. In this paper, we propose a novel\nconformal loss function that approximates the traditionally two-step CP\napproach in a single step. By evaluating and penalising deviations from the\nstringent expected CP output distribution, a Deep Learning model may learn the\ndirect relationship between the input data and the conformal p-values. We carry\nout a comprehensive empirical evaluation to show our novel loss function's\ncompetitiveness for seven binary and multi-class prediction tasks on five\nbenchmark datasets. On the same datasets, our approach achieves significant\ntraining time reductions up to 86% compared to Aggregated Conformal Prediction\n(ACP), while maintaining comparable approximate validity and predictive\nefficiency.\n", "date": "2022-07-25", "categories": []} {"id": "2307.16360", "title": "Probabilistically robust conformal prediction", "abstract": " Conformal prediction (CP) is a framework to quantify uncertainty of machine\nlearning classifiers including deep neural networks. Given a testing example\nand a trained classifier, CP produces a prediction set of candidate labels with\na user-specified coverage (i.e., true class label is contained with high\nprobability). Almost all the existing work on CP assumes clean testing data and\nthere is not much known about the robustness of CP algorithms w.r.t\nnatural/adversarial perturbations to testing examples. This paper studies the\nproblem of probabilistically robust conformal prediction (PRCP) which ensures\nrobustness to most perturbations around clean input examples. PRCP generalizes\nthe standard CP (cannot handle perturbations) and adversarially robust CP\n(ensures robustness w.r.t worst-case perturbations) to achieve better\ntrade-offs between nominal performance and robustness. We propose a novel\nadaptive PRCP (aPRCP) algorithm to achieve probabilistically robust coverage.\nThe key idea behind aPRCP is to determine two parallel thresholds, one for data\nsamples and another one for the perturbations on data (aka\n\"quantile-of-quantile\" design). We provide theoretical analysis to show that\naPRCP algorithm achieves robust coverage. Our experiments on CIFAR-10,\nCIFAR-100, and ImageNet datasets using deep neural networks demonstrate that\naPRCP achieves better trade-offs than state-of-the-art CP and adversarially\nrobust CP algorithms.\n", "date": "2023-07-31", "categories": []} {"id": "2307.16895", "title": "Conformal PID Control for Time Series Prediction", "abstract": " We study the problem of uncertainty quantification for time series\nprediction, with the goal of providing easy-to-use algorithms with formal\nguarantees. The algorithms we present build upon ideas from conformal\nprediction and control theory, are able to prospectively model conformal scores\nin an online setting, and adapt to the presence of systematic errors due to\nseasonality, trends, and general distribution shifts. Our theory both\nsimplifies and strengthens existing analyses in online conformal prediction.\nExperiments on 4-week-ahead forecasting of statewide COVID-19 death counts in\nthe U.S. show an improvement in coverage over the ensemble forecaster used in\nofficial CDC communications. We also run experiments on predicting electricity\ndemand, market returns, and temperature using autoregressive, Theta, Prophet,\nand Transformer models. We provide an extendable codebase for testing our\nmethods and for the integration of new algorithms, data sets, and forecasting\nrules.\n", "date": "2023-07-31", "categories": []} {"id": "2307.05109", "title": "Conformalization of Sparse Generalized Linear Models", "abstract": " Given a sequence of observable variables $\\{(x_1, y_1), \\ldots, (x_n,\ny_n)\\}$, the conformal prediction method estimates a confidence set for\n$y_{n+1}$ given $x_{n+1}$ that is valid for any finite sample size by merely\nassuming that the joint distribution of the data is permutation invariant.\nAlthough attractive, computing such a set is computationally infeasible in most\nregression problems. Indeed, in these cases, the unknown variable $y_{n+1}$ can\ntake an infinite number of possible candidate values, and generating conformal\nsets requires retraining a predictive model for each candidate. In this paper,\nwe focus on a sparse linear model with only a subset of variables for\nprediction and use numerical continuation techniques to approximate the\nsolution path efficiently. The critical property we exploit is that the set of\nselected variables is invariant under a small perturbation of the input data.\nTherefore, it is sufficient to enumerate and refit the model only at the change\npoints of the set of active features and smoothly interpolate the rest of the\nsolution via a Predictor-Corrector mechanism. We show how our path-following\nalgorithm accurately approximates conformal prediction sets and illustrate its\nperformance using synthetic and real data examples.\n", "date": "2023-07-11", "categories": []} {"id": "2305.18404", "title": "Conformal Prediction with Large Language Models for Multi-Choice\n Question Answering", "abstract": " As large language models continue to be widely developed, robust uncertainty\nquantification techniques will become crucial for their safe deployment in\nhigh-stakes scenarios. In this work, we explore how conformal prediction can be\nused to provide uncertainty quantification in language models for the specific\ntask of multiple-choice question-answering. We find that the uncertainty\nestimates from conformal prediction are tightly correlated with prediction\naccuracy. This observation can be useful for downstream applications such as\nselective classification and filtering out low-quality predictions. We also\ninvestigate the exchangeability assumption required by conformal prediction to\nout-of-subject questions, which may be a more realistic scenario for many\npractical applications. Our work contributes towards more trustworthy and\nreliable usage of large language models in safety-critical situations, where\nrobust guarantees of error rate are required.\n", "date": "2023-05-28", "categories": []} {"id": "2307.01088", "title": "Empirically Validating Conformal Prediction on Modern Vision\n Architectures Under Distribution Shift and Long-tailed Data", "abstract": " Conformal prediction has emerged as a rigorous means of providing deep\nlearning models with reliable uncertainty estimates and safety guarantees. Yet,\nits performance is known to degrade under distribution shift and long-tailed\nclass distributions, which are often present in real world applications. Here,\nwe characterize the performance of several post-hoc and training-based\nconformal prediction methods under these settings, providing the first\nempirical evaluation on large-scale datasets and models. We show that across\nnumerous conformal methods and neural network families, performance greatly\ndegrades under distribution shifts violating safety guarantees. Similarly, we\nshow that in long-tailed settings the guarantees are frequently violated on\nmany classes. Understanding the limitations of these methods is necessary for\ndeployment in real world and safety-critical applications.\n", "date": "2023-07-03", "categories": []} {"id": "2306.17169", "title": "Enterprise Disk Drive Scrubbing Based on Mondrian Conformal Predictors", "abstract": " Disk scrubbing is a process aimed at resolving read errors on disks by\nreading data from the disk. However, scrubbing the entire storage array at once\ncan adversely impact system performance, particularly during periods of high\ninput/output operations. Additionally, the continuous reading of data from\ndisks when scrubbing can result in wear and tear, especially on larger capacity\ndisks, due to the significant time and energy consumption involved. To address\nthese issues, we propose a selective disk scrubbing method that enhances the\noverall reliability and power efficiency in data centers. Our method employs a\nMachine Learning model based on Mondrian Conformal prediction to identify\nspecific disks for scrubbing, by proactively predicting the health status of\neach disk in the storage pool, forecasting n-days in advance, and using an\nopen-source dataset. For disks predicted as non-healthy, we mark them for\nreplacement without further action. For healthy drives, we create a set and\nquantify their relative health across the entire storage pool based on the\npredictor's confidence. This enables us to prioritize selective scrubbing for\ndrives with established scrubbing frequency based on the scrub cycle. The\nmethod we propose provides an efficient and dependable solution for managing\nenterprise disk drives. By scrubbing just 22.7% of the total storage disks, we\ncan achieve optimized energy consumption and reduce the carbon footprint of the\ndata center.\n", "date": "2023-06-01", "categories": []} {"id": "2301.11556", "title": "Conformal inference is (almost) free for neural networks trained with\n early stopping", "abstract": " Early stopping based on hold-out data is a popular regularization technique\ndesigned to mitigate overfitting and increase the predictive accuracy of neural\nnetworks. Models trained with early stopping often provide relatively accurate\npredictions, but they generally still lack precise statistical guarantees\nunless they are further calibrated using independent hold-out data. This paper\naddresses the above limitation with conformalized early stopping: a novel\nmethod that combines early stopping with conformal calibration while\nefficiently recycling the same hold-out data. This leads to models that are\nboth accurate and able to provide exact predictive inferences without multiple\ndata splits nor overly conservative adjustments. Practical implementations are\ndeveloped for different learning tasks -- outlier detection, multi-class\nclassification, regression -- and their competitive performance is demonstrated\non real data.\n", "date": "2023-01-27", "categories": []} {"id": "2302.08703", "title": "PAC Prediction Sets for Large Language Models of Code", "abstract": " Prediction sets have recently been shown to be a promising strategy for\nquantifying the uncertainty of deep neural networks in a way that provides\ntheoretical guarantees. However, existing techniques have largely targeted\nsettings where the space of labels is simple, so prediction sets can be\narbitrary subsets of labels. For structured prediction problems where the space\nof labels is exponential in size, even prediction sets containing a small\nfraction of all labels can be exponentially large. In the context of code\ngeneration, we propose a solution that considers a restricted set of prediction\nsets that can compactly be represented as partial programs, which are programs\nwith portions replaced with holes. Given a trained code generation model, our\nalgorithm leverages a programming language's abstract syntax tree to generate a\nset of programs such that the correct program is in the set with\nhigh-confidence. Valuable applications of our algorithm include a Codex-style\ncode generator with holes in uncertain parts of the generated code, which\nprovides a partial program with theoretical guarantees. We evaluate our\napproach on PICARD (a T5 model for SQL semantic parsing) and Codex (a GPT model\nfor over a dozen programming languages, including Python), demonstrating that\nour approach generates compact PAC prediction sets. This is the first research\ncontribution that generates PAC prediction sets for generative code models.\n", "date": "2023-02-17", "categories": []} {"id": "2306.06221", "title": "Conformalizing Machine Translation Evaluation", "abstract": " Several uncertainty estimation methods have been recently proposed for\nmachine translation evaluation. While these methods can provide a useful\nindication of when not to trust model predictions, we show in this paper that\nthe majority of them tend to underestimate model uncertainty, and as a result\nthey often produce misleading confidence intervals that do not cover the ground\ntruth. We propose as an alternative the use of conformal prediction, a\ndistribution-free method to obtain confidence intervals with a theoretically\nestablished guarantee on coverage. First, we demonstrate that split conformal\nprediction can ``correct'' the confidence intervals of previous methods to\nyield a desired coverage level. Then, we highlight biases in estimated\nconfidence intervals, both in terms of the translation language pairs and the\nquality of translations. We apply conditional conformal prediction techniques\nto obtain calibration subsets for each data subgroup, leading to equalized\ncoverage.\n", "date": "2023-06-09", "categories": []} {"id": "2306.04648", "title": "On training locally adaptive CP", "abstract": " We address the problem of making Conformal Prediction (CP) intervals locally\nadaptive. Most existing methods focus on approximating the object-conditional\nvalidity of the intervals by partitioning or re-weighting the calibration set.\nOur strategy is new and conceptually different. Instead of re-weighting the\ncalibration data, we redefine the conformity measure through a trainable change\nof variables, $A \\to \\phi_X(A)$, that depends explicitly on the object\nattributes, $X$. Under certain conditions and if $\\phi_X$ is monotonic in $A$\nfor any $X$, the transformations produce prediction intervals that are\nguaranteed to be marginally valid and have $X$-dependent sizes. We describe how\nto parameterize and train $\\phi_X$ to maximize the interval efficiency.\nContrary to other CP-aware training methods, the objective function is smooth\nand can be minimized through standard gradient methods without approximations.\n", "date": "2023-06-05", "categories": []} {"id": "2305.11640", "title": "Distribution-Free Matrix Prediction Under Arbitrary Missing Pattern", "abstract": " This paper studies the open problem of conformalized entry prediction in a\nrow/column-exchangeable matrix. The matrix setting presents novel and unique\nchallenges, but there exists little work on this interesting topic. We\nmeticulously define the problem, differentiate it from closely related\nproblems, and rigorously delineate the boundary between achievable and\nimpossible goals. We then propose two practical algorithms. The first method\nprovides a fast emulation of the full conformal prediction, while the second\nmethod leverages the technique of algorithmic stability for acceleration. Both\nmethods are computationally efficient and can effectively safeguard coverage\nvalidity in presence of arbitrary missing pattern. Further, we quantify the\nimpact of missingness on prediction accuracy and establish fundamental limit\nresults. Empirical evidence from synthetic and real-world data sets\ncorroborates the superior performance of our proposed methods.\n", "date": "2023-05-19", "categories": []} {"id": "2210.04166", "title": "Test-time Recalibration of Conformal Predictors Under Distribution Shift\n Based on Unlabeled Examples", "abstract": " Modern image classifiers are very accurate, but the predictions come without\nuncertainty estimates. Conformal predictors provide uncertainty estimates by\ncomputing a set of classes containing the correct class with a user-specified\nprobability based on the classifier's probability estimates. To provide such\nsets, conformal predictors often estimate a cutoff threshold for the\nprobability estimates based on a calibration set. Conformal predictors\nguarantee reliability only when the calibration set is from the same\ndistribution as the test set. Therefore, conformal predictors need to be\nrecalibrated for new distributions. However, in practice, labeled data from new\ndistributions is rarely available, making calibration infeasible. In this work,\nwe consider the problem of predicting the cutoff threshold for a new\ndistribution based on unlabeled examples. While it is impossible in general to\nguarantee reliability when calibrating based on unlabeled examples, we propose\na method that provides excellent uncertainty estimates under natural\ndistribution shifts, and provably works for a specific model of a distribution\nshift.\n", "date": "2022-10-09", "categories": []} {"id": "2306.02732", "title": "Conformal Prediction with Missing Values", "abstract": " Conformal prediction is a theoretically grounded framework for constructing\npredictive intervals. We study conformal prediction with missing values in the\ncovariates -- a setting that brings new challenges to uncertainty\nquantification. We first show that the marginal coverage guarantee of conformal\nprediction holds on imputed data for any missingness distribution and almost\nall imputation functions. However, we emphasize that the average coverage\nvaries depending on the pattern of missing values: conformal methods tend to\nconstruct prediction intervals that under-cover the response conditionally to\nsome missing patterns. This motivates our novel generalized conformalized\nquantile regression framework, missing data augmentation, which yields\nprediction intervals that are valid conditionally to the patterns of missing\nvalues, despite their exponential number. We then show that a universally\nconsistent quantile regression algorithm trained on the imputed data is Bayes\noptimal for the pinball risk, thus achieving valid coverage conditionally to\nany given data point. Moreover, we examine the case of a linear model, which\ndemonstrates the importance of our proposal in overcoming the\nheteroskedasticity induced by missing values. Using synthetic and data from\ncritical care, we corroborate our theory and report improved performance of our\nmethods.\n", "date": "2023-06-05", "categories": []} {"id": "2306.01191", "title": "Conformal Prediction with Partially Labeled Data", "abstract": " While the predictions produced by conformal prediction are set-valued, the\ndata used for training and calibration is supposed to be precise. In the\nsetting of superset learning or learning from partial labels, a variant of\nweakly supervised learning, it is exactly the other way around: training data\nis possibly imprecise (set-valued), but the model induced from this data yields\nprecise predictions. In this paper, we combine the two settings by making\nconformal prediction amenable to set-valued training data. We propose a\ngeneralization of the conformal prediction procedure that can be applied to\nset-valued training and calibration data. We prove the validity of the proposed\nmethod and present experimental studies in which it compares favorably to\nnatural baselines.\n", "date": "2023-06-01", "categories": []} {"id": "2306.01549", "title": "Evaluating Machine Translation Quality with Conformal Predictive\n Distributions", "abstract": " This paper presents a new approach for assessing uncertainty in machine\ntranslation by simultaneously evaluating translation quality and providing a\nreliable confidence score. Our approach utilizes conformal predictive\ndistributions to produce prediction intervals with guaranteed coverage, meaning\nthat for any given significance level $\\epsilon$, we can expect the true\nquality score of a translation to fall out of the interval at a rate of\n$1-\\epsilon$. In this paper, we demonstrate how our method outperforms a\nsimple, but effective baseline on six different language pairs in terms of\ncoverage and sharpness. Furthermore, we validate that our approach requires the\ndata exchangeability assumption to hold for optimal performance.\n", "date": "2023-06-02", "categories": []} {"id": "2305.17564", "title": "Federated Conformal Predictors for Distributed Uncertainty\n Quantification", "abstract": " Conformal prediction is emerging as a popular paradigm for providing rigorous\nuncertainty quantification in machine learning since it can be easily applied\nas a post-processing step to already trained models. In this paper, we extend\nconformal prediction to the federated learning setting. The main challenge we\nface is data heterogeneity across the clients - this violates the fundamental\ntenet of exchangeability required for conformal prediction. We propose a weaker\nnotion of partial exchangeability, better suited to the FL setting, and use it\nto develop the Federated Conformal Prediction (FCP) framework. We show FCP\nenjoys rigorous theoretical guarantees and excellent empirical performance on\nseveral computer vision and medical imaging datasets. Our results demonstrate a\npractical approach to incorporating meaningful uncertainty quantification in\ndistributed and heterogeneous environments. We provide code used in our\nexperiments https://github.com/clu5/federated-conformal.\n", "date": "2023-05-27", "categories": []} {"id": "2303.12695", "title": "Adaptive Conformal Prediction by Reweighting Nonconformity Score", "abstract": " Despite attractive theoretical guarantees and practical successes, Predictive\nInterval (PI) given by Conformal Prediction (CP) may not reflect the\nuncertainty of a given model. This limitation arises from CP methods using a\nconstant correction for all test points, disregarding their individual\nuncertainties, to ensure coverage properties. To address this issue, we propose\nusing a Quantile Regression Forest (QRF) to learn the distribution of\nnonconformity scores and utilizing the QRF's weights to assign more importance\nto samples with residuals similar to the test point. This approach results in\nPI lengths that are more aligned with the model's uncertainty. In addition, the\nweights learnt by the QRF provide a partition of the features space, allowing\nfor more efficient computations and improved adaptiveness of the PI through\ngroupwise conformalization. Our approach enjoys an assumption-free finite\nsample marginal and training-conditional coverage, and under suitable\nassumptions, it also ensures conditional coverage. Our methods work for any\nnonconformity score and are available as a Python package. We conduct\nexperiments on simulated and real-world data that demonstrate significant\nimprovements compared to existing methods.\n", "date": "2023-03-22", "categories": []} {"id": "2305.19901", "title": "Adaptive Conformal Regression with Jackknife+ Rescaled Scores", "abstract": " Conformal regression provides prediction intervals with global coverage\nguarantees, but often fails to capture local error distributions, leading to\nnon-homogeneous coverage. We address this with a new adaptive method based on\nrescaling conformal scores with an estimate of local score distribution,\ninspired by the Jackknife+ method, which enables the use of calibration data in\nconformal scores without breaking calibration-test exchangeability. Our\napproach ensures formal global coverage guarantees and is supported by new\ntheoretical results on local coverage, including an a posteriori bound on any\ncalibration score. The strength of our approach lies in achieving local\ncoverage without sacrificing calibration set size, improving the applicability\nof conformal prediction intervals in various settings. As a result, our method\nprovides prediction intervals that outperform previous methods, particularly in\nthe low-data regime, making it especially relevant for real-world applications\nsuch as healthcare and biomedical domains where uncertainty needs to be\nquantified accurately despite low sample data.\n", "date": "2023-05-31", "categories": []} {"id": "2212.03463", "title": "Sequential Predictive Conformal Inference for Time Series", "abstract": " We present a new distribution-free conformal prediction algorithm for\nsequential data (e.g., time series), called the \\textit{sequential predictive\nconformal inference} (\\texttt{SPCI}). We specifically account for the nature\nthat time series data are non-exchangeable, and thus many existing conformal\nprediction algorithms are not applicable. The main idea is to adaptively\nre-estimate the conditional quantile of non-conformity scores (e.g., prediction\nresiduals), upon exploiting the temporal dependence among them. More precisely,\nwe cast the problem of conformal prediction interval as predicting the quantile\nof a future residual, given a user-specified point prediction algorithm.\nTheoretically, we establish asymptotic valid conditional coverage upon\nextending consistency analyses in quantile regression. Using simulation and\nreal-data experiments, we demonstrate a significant reduction in interval width\nof \\texttt{SPCI} compared to other existing methods under the desired empirical\ncoverage.\n", "date": "2022-12-07", "categories": []} {"id": "2210.02271", "title": "Extending Conformal Prediction to Hidden Markov Models with Exact\n Validity via de Finetti's Theorem for Markov Chains", "abstract": " Conformal prediction is a widely used method to quantify the uncertainty of a\nclassifier under the assumption of exchangeability (e.g., IID data). We\ngeneralize conformal prediction to the Hidden Markov Model (HMM) framework\nwhere the assumption of exchangeability is not valid. The key idea of the\nproposed method is to partition the non-exchangeable Markovian data from the\nHMM into exchangeable blocks by exploiting the de Finetti's Theorem for Markov\nChains discovered by Diaconis and Freedman (1980). The permutations of the\nexchangeable blocks are viewed as randomizations of the observed Markovian data\nfrom the HMM. The proposed method provably retains all desirable theoretical\nguarantees offered by the classical conformal prediction framework in both\nexchangeable and Markovian settings. In particular, while the lack of\nexchangeability introduced by Markovian samples constitutes a violation of a\ncrucial assumption for classical conformal prediction, the proposed method\nviews it as an advantage that can be exploited to improve the performance\nfurther. Detailed numerical and empirical results that complement the\ntheoretical conclusions are provided to illustrate the practical feasibility of\nthe proposed method.\n", "date": "2022-10-05", "categories": []} {"id": "2302.00839", "title": "Fast Online Value-Maximizing Prediction Sets with Conformal Cost Control", "abstract": " Many real-world multi-label prediction problems involve set-valued\npredictions that must satisfy specific requirements dictated by downstream\nusage. We focus on a typical scenario where such requirements, separately\nencoding $\\textit{value}$ and $\\textit{cost}$, compete with each other. For\ninstance, a hospital might expect a smart diagnosis system to capture as many\nsevere, often co-morbid, diseases as possible (the value), while maintaining\nstrict control over incorrect predictions (the cost). We present a general\npipeline, dubbed as FavMac, to maximize the value while controlling the cost in\nsuch scenarios. FavMac can be combined with almost any multi-label classifier,\naffording distribution-free theoretical guarantees on cost control. Moreover,\nunlike prior works, it can handle real-world large-scale applications via a\ncarefully designed online update mechanism, which is of independent interest.\nOur methodological and theoretical contributions are supported by experiments\non several healthcare tasks and synthetic datasets - FavMac furnishes higher\nvalue compared with several variants and baselines while maintaining strict\ncost control. Our code is available at https://github.com/zlin7/FavMac\n", "date": "2023-02-02", "categories": []} {"id": "2210.00173", "title": "Predictive Inference with Feature Conformal Prediction", "abstract": " Conformal prediction is a distribution-free technique for establishing valid\nprediction intervals. Although conventionally people conduct conformal\nprediction in the output space, this is not the only possibility. In this\npaper, we propose feature conformal prediction, which extends the scope of\nconformal prediction to semantic feature spaces by leveraging the inductive\nbias of deep representation learning. From a theoretical perspective, we\ndemonstrate that feature conformal prediction provably outperforms regular\nconformal prediction under mild assumptions. Our approach could be combined\nwith not only vanilla conformal prediction, but also other adaptive conformal\nprediction methods. Apart from experiments on existing predictive inference\nbenchmarks, we also demonstrate the state-of-the-art performance of the\nproposed methods on large-scale tasks such as ImageNet classification and\nCityscapes image segmentation.The code is available at\n\\url{https://github.com/AlvinWen428/FeatureCP}.\n", "date": "2022-10-01", "categories": []} {"id": "2304.01426", "title": "Conformalized Unconditional Quantile Regression", "abstract": " We develop a predictive inference procedure that combines conformal\nprediction (CP) with unconditional quantile regression (QR) -- a commonly used\ntool in econometrics that involves regressing the recentered influence function\n(RIF) of the quantile functional over input covariates. Unlike the more\nwidely-known conditional QR, unconditional QR explicitly captures the impact of\nchanges in covariate distribution on the quantiles of the marginal distribution\nof outcomes. Leveraging this property, our procedure issues adaptive predictive\nintervals with localized frequentist coverage guarantees. It operates by\nfitting a machine learning model for the RIFs using training data, and then\napplying the CP procedure for any test covariate with respect to a\n``hypothetical'' covariate distribution localized around the new instance.\nExperiments show that our procedure is adaptive to heteroscedasticity, provides\ntransparent coverage guarantees that are relevant to the test instance at hand,\nand performs competitively with existing methods in terms of efficiency.\n", "date": "2023-04-04", "categories": []} {"id": "2304.00970", "title": "Development and Evaluation of Conformal Prediction Methods for QSAR", "abstract": " The quantitative structure-activity relationship (QSAR) regression model is a\ncommonly used technique for predicting biological activities of compounds using\ntheir molecular descriptors. Predictions from QSAR models can help, for\nexample, to optimize molecular structure; prioritize compounds for further\nexperimental testing; and estimate their toxicity. In addition to the accurate\nestimation of the activity, it is highly desirable to obtain some estimate of\nthe uncertainty associated with the prediction, e.g., calculate a prediction\ninterval (PI) containing the true molecular activity with a pre-specified\nprobability, say 70%, 90% or 95%. The challenge is that most machine learning\n(ML) algorithms that achieve superior predictive performance require some\nadd-on methods for estimating uncertainty of their prediction. The development\nof these algorithms is an active area of research by statistical and ML\ncommunities but their implementation for QSAR modeling remains limited.\nConformal prediction (CP) is a promising approach. It is agnostic to the\nprediction algorithm and can produce valid prediction intervals under some weak\nassumptions on the data distribution. We proposed computationally efficient CP\nalgorithms tailored to the most advanced ML models, including Deep Neural\nNetworks and Gradient Boosting Machines. The validity and efficiency of\nproposed conformal predictors are demonstrated on a diverse collection of QSAR\ndatasets as well as simulation studies.\n", "date": "2023-04-03", "categories": []} {"id": "2303.03995", "title": "Group conditional validity via multi-group learning", "abstract": " We consider the problem of distribution-free conformal prediction and the\ncriterion of group conditional validity. This criterion is motivated by many\npractical scenarios including hidden stratification and group fairness.\nExisting methods achieve such guarantees under either restrictive grouping\nstructure or distributional assumptions, or they are overly-conservative under\nheteroskedastic noise. We propose a simple reduction to the problem of\nachieving validity guarantees for individual populations by leveraging\nalgorithms for a problem called multi-group learning. This allows us to port\ntheoretical guarantees from multi-group learning to obtain obtain sample\ncomplexity guarantees for conformal prediction. We also provide a new algorithm\nfor multi-group learning for groups with hierarchical structure. Using this\nalgorithm in our reduction leads to improved sample complexity guarantees with\na simpler predictor structure.\n", "date": "2023-03-07", "categories": []} {"id": "2303.10694", "title": "Improving Uncertainty Quantification of Deep Classifiers via\n Neighborhood Conformal Prediction: Novel Algorithm and Theoretical Analysis", "abstract": " Safe deployment of deep neural networks in high-stake real-world applications\nrequires theoretically sound uncertainty quantification. Conformal prediction\n(CP) is a principled framework for uncertainty quantification of deep models in\nthe form of prediction set for classification tasks with a user-specified\ncoverage (i.e., true class label is contained with high probability). This\npaper proposes a novel algorithm referred to as Neighborhood Conformal\nPrediction (NCP) to improve the efficiency of uncertainty quantification from\nCP for deep classifiers (i.e., reduce prediction set size). The key idea behind\nNCP is to use the learned representation of the neural network to identify k\nnearest-neighbors calibration examples for a given testing input and assign\nthem importance weights proportional to their distance to create adaptive\nprediction sets. We theoretically show that if the learned data representation\nof the neural network satisfies some mild conditions, NCP will produce smaller\nprediction sets than traditional CP algorithms. Our comprehensive experiments\non CIFAR-10, CIFAR-100, and ImageNet datasets using diverse deep neural\nnetworks strongly demonstrate that NCP leads to significant reduction in\nprediction set size over prior CP methods.\n", "date": "2023-03-19", "categories": []} {"id": "2211.09330", "title": "ACon$^2$: Adaptive Conformal Consensus for Provable Blockchain Oracles", "abstract": " Blockchains with smart contracts are distributed ledger systems that achieve\nblock-state consistency among distributed nodes by only allowing deterministic\noperations of smart contracts. However, the power of smart contracts is enabled\nby interacting with stochastic off-chain data, which in turn opens the\npossibility to undermine the block-state consistency. To address this issue, an\noracle smart contract is used to provide a single consistent source of external\ndata; but, simultaneously, this introduces a single point of failure, which is\ncalled the oracle problem. To address the oracle problem, we propose an\nadaptive conformal consensus (ACon$^2$) algorithm that derives a consensus set\nof data from multiple oracle contracts via the recent advance in online\nuncertainty quantification learning. Interesting, the consensus set provides a\ndesired correctness guarantee under distribution shift and Byzantine\nadversaries. We demonstrate the efficacy of the proposed algorithm on two price\ndatasets and an Ethereum case study. In particular, the Solidity implementation\nof the proposed algorithm shows the potential practicality of the proposed\nalgorithm, implying that online machine learning algorithms are applicable to\naddress security issues in blockchains.\n", "date": "2022-11-17", "categories": []} {"id": "2303.02207", "title": "Lightweight, Uncertainty-Aware Conformalized Visual Odometry", "abstract": " Data-driven visual odometry (VO) is a critical subroutine for autonomous edge\nrobotics, and recent progress in the field has produced highly accurate point\npredictions in complex environments. However, emerging autonomous edge robotics\ndevices like insect-scale drones and surgical robots lack a computationally\nefficient framework to estimate VO's predictive uncertainties. Meanwhile, as\nedge robotics continue to proliferate into mission-critical application spaces,\nawareness of model's the predictive uncertainties has become crucial for\nrisk-aware decision-making. This paper addresses this challenge by presenting a\nnovel, lightweight, and statistically robust framework that leverages conformal\ninference (CI) to extract VO's uncertainty bands. Our approach represents the\nuncertainties using flexible, adaptable, and adjustable prediction intervals\nthat, on average, guarantee the inclusion of the ground truth across all\ndegrees of freedom (DOF) of pose estimation. We discuss the architectures of\ngenerative deep neural networks for estimating multivariate uncertainty bands\nalong with point (mean) prediction. We also present techniques to improve the\nuncertainty estimation accuracy, such as leveraging Monte Carlo dropout\n(MC-dropout) for data augmentation. Finally, we propose a novel training loss\nfunction that combines interval scoring and calibration loss with traditional\ntraining metrics--mean-squared error and KL-divergence--to improve\nuncertainty-aware learning. Our simulation results demonstrate that the\npresented framework consistently captures true uncertainty in pose estimations\nacross different datasets, estimation models, and applied noise types,\nindicating its wide applicability.\n", "date": "2023-03-03", "categories": []} {"id": "2302.12238", "title": "Improving Adaptive Conformal Prediction Using Self-Supervised Learning", "abstract": " Conformal prediction is a powerful distribution-free tool for uncertainty\nquantification, establishing valid prediction intervals with finite-sample\nguarantees. To produce valid intervals which are also adaptive to the\ndifficulty of each instance, a common approach is to compute normalized\nnonconformity scores on a separate calibration set. Self-supervised learning\nhas been effectively utilized in many domains to learn general representations\nfor downstream predictors. However, the use of self-supervision beyond model\npretraining and representation learning has been largely unexplored. In this\nwork, we investigate how self-supervised pretext tasks can improve the quality\nof the conformal regressors, specifically by improving the adaptability of\nconformal intervals. We train an auxiliary model with a self-supervised pretext\ntask on top of an existing predictive model and use the self-supervised error\nas an additional feature to estimate nonconformity scores. We empirically\ndemonstrate the benefit of the additional information using both synthetic and\nreal data on the efficiency (width), deficit, and excess of conformal\nprediction intervals.\n", "date": "2023-02-23", "categories": []} {"id": "2206.13092", "title": "Split Localized Conformal Prediction", "abstract": " Conformal prediction is a simple and powerful tool that can quantify\nuncertainty without any distributional assumptions. Many existing methods only\naddress the average coverage guarantee, which is not ideal compared to the\nstronger conditional coverage guarantee. Existing methods of approximating\nconditional coverage require additional models or time effort, which makes them\nnot easy to scale. In this paper, we propose a modified non-conformity score by\nleveraging the local approximation of the conditional distribution using kernel\ndensity estimation. The modified score inherits the spirit of split conformal\nmethods, which is simple and efficient and can scale to high dimensional\nsettings. We also proposed a unified framework that brings together our method\nand several state-of-the-art. We perform extensive empirical evaluations:\nresults measured by both average and conditional coverage confirm the advantage\nof our method.\n", "date": "2022-06-27", "categories": []} {"id": "2302.07869", "title": "Improved Online Conformal Prediction via Strongly Adaptive Online\n Learning", "abstract": " We study the problem of uncertainty quantification via prediction sets, in an\nonline setting where the data distribution may vary arbitrarily over time.\nRecent work develops online conformal prediction techniques that leverage\nregret minimization algorithms from the online learning literature to learn\nprediction sets with approximately valid coverage and small regret. However,\nstandard regret minimization could be insufficient for handling changing\nenvironments, where performance guarantees may be desired not only over the\nfull time horizon but also in all (sub-)intervals of time. We develop new\nonline conformal prediction methods that minimize the strongly adaptive regret,\nwhich measures the worst-case regret over all intervals of a fixed length. We\nprove that our methods achieve near-optimal strongly adaptive regret for all\ninterval lengths simultaneously, and approximately valid coverage. Experiments\nshow that our methods consistently obtain better coverage and smaller\nprediction sets than existing methods on real-world tasks, such as time series\nforecasting and image classification under distribution shift.\n", "date": "2023-02-15", "categories": []} {"id": "2302.05304", "title": "From Group-Differences to Single-Subject Probability: Conformal\n Prediction-based Uncertainty Estimation for Brain-Age Modeling", "abstract": " The brain-age gap is one of the most investigated risk markers for brain\nchanges across disorders. While the field is progressing towards large-scale\nmodels, recently incorporating uncertainty estimates, no model to date provides\nthe single-subject risk assessment capability essential for clinical\napplication. In order to enable the clinical use of brain-age as a biomarker,\nwe here combine uncertainty-aware deep Neural Networks with conformal\nprediction theory. This approach provides statistical guarantees with respect\nto single-subject uncertainty estimates and allows for the calculation of an\nindividual's probability for accelerated brain-aging. Building on this, we show\nempirically in a sample of N=16,794 participants that 1. a lower or comparable\nerror as state-of-the-art, large-scale brain-age models, 2. the statistical\nguarantees regarding single-subject uncertainty estimation indeed hold for\nevery participant, and 3. that the higher individual probabilities of\naccelerated brain-aging derived from our model are associated with Alzheimer's\nDisease, Bipolar Disorder and Major Depressive Disorder.\n", "date": "2023-02-10", "categories": []} {"id": "2302.00872", "title": "Reliable Prediction Intervals with Directly Optimized Inductive\n Conformal Regression for Deep Learning", "abstract": " By generating prediction intervals (PIs) to quantify the uncertainty of each\nprediction in deep learning regression, the risk of wrong predictions can be\neffectively controlled. High-quality PIs need to be as narrow as possible,\nwhilst covering a preset proportion of real labels. At present, many approaches\nto improve the quality of PIs can effectively reduce the width of PIs, but they\ndo not ensure that enough real labels are captured. Inductive Conformal\nPredictor (ICP) is an algorithm that can generate effective PIs which is\ntheoretically guaranteed to cover a preset proportion of data. However,\ntypically ICP is not directly optimized to yield minimal PI width. However, in\nthis study, we use Directly Optimized Inductive Conformal Regression (DOICR)\nthat takes only the average width of PIs as the loss function and increases the\nquality of PIs through an optimized scheme under the validity condition that\nsufficient real labels are captured in the PIs. Benchmark experiments show that\nDOICR outperforms current state-of-the-art algorithms for regression problems\nusing underlying Deep Neural Network structures for both tabular and image\ndata.\n", "date": "2023-02-02", "categories": []} {"id": "2212.07775", "title": "Calibrating AI Models for Wireless Communications via Conformal\n Prediction", "abstract": " When used in complex engineered systems, such as communication networks,\nartificial intelligence (AI) models should be not only as accurate as possible,\nbut also well calibrated. A well-calibrated AI model is one that can reliably\nquantify the uncertainty of its decisions, assigning high confidence levels to\ndecisions that are likely to be correct and low confidence levels to decisions\nthat are likely to be erroneous. This paper investigates the application of\nconformal prediction as a general framework to obtain AI models that produce\ndecisions with formal calibration guarantees. Conformal prediction transforms\nprobabilistic predictors into set predictors that are guaranteed to contain the\ncorrect answer with a probability chosen by the designer. Such formal\ncalibration guarantees hold irrespective of the true, unknown, distribution\nunderlying the generation of the variables of interest, and can be defined in\nterms of ensemble or time-averaged probabilities. In this paper, conformal\nprediction is applied for the first time to the design of AI for communication\nsystems in conjunction to both frequentist and Bayesian learning, focusing on\ndemodulation, modulation classification, and channel prediction.\n", "date": "2022-12-15", "categories": []} {"id": "2211.16238", "title": "A Cross-Conformal Predictor for Multi-label Classification", "abstract": " Unlike the typical classification setting where each instance is associated\nwith a single class, in multi-label learning each instance is associated with\nmultiple classes simultaneously. Therefore the learning task in this setting is\nto predict the subset of classes to which each instance belongs. This work\nexamines the application of a recently developed framework called Conformal\nPrediction (CP) to the multi-label learning setting. CP complements the\npredictions of machine learning algorithms with reliable measures of\nconfidence. As a result the proposed approach instead of just predicting the\nmost likely subset of classes for a new unseen instance, also indicates the\nlikelihood of each predicted subset being correct. This additional information\nis especially valuable in the multi-label setting where the overall uncertainty\nis extremely high.\n", "date": "2022-11-29", "categories": []} {"id": "2210.15805", "title": "Towards Reliable Zero Shot Classification in Self-Supervised Models with\n Conformal Prediction", "abstract": " Self-supervised models trained with a contrastive loss such as CLIP have\nshown to be very powerful in zero-shot classification settings. However, to be\nused as a zero-shot classifier these models require the user to provide new\ncaptions over a fixed set of labels at test time. In many settings, it is hard\nor impossible to know if a new query caption is compatible with the source\ncaptions used to train the model. We address these limitations by framing the\nzero-shot classification task as an outlier detection problem and develop a\nconformal prediction procedure to assess when a given test caption may be\nreliably used. On a real-world medical example, we show that our proposed\nconformal procedure improves the reliability of CLIP-style models in the\nzero-shot classification setting, and we provide an empirical analysis of the\nfactors that may affect its performance.\n", "date": "2022-10-27", "categories": []} {"id": "2210.14735", "title": "Distribution-Free Finite-Sample Guarantees and Split Conformal\n Prediction", "abstract": " Modern black-box predictive models are often accompanied by weak performance\nguarantees that only hold asymptotically in the size of the dataset or require\nstrong parametric assumptions. In response to this, split conformal prediction\nrepresents a promising avenue to obtain finite-sample guarantees under minimal\ndistribution-free assumptions. Although prediction set validity most often\nconcerns marginal coverage, we explore the related but different guarantee of\ntolerance regions, reformulating known results in the language of nested\nprediction sets and extending on the duality between marginal coverage and\ntolerance regions. Furthermore, we highlight the connection between split\nconformal prediction and classical tolerance predictors developed in the 1940s,\nas well as recent developments in distribution-free risk control. One result\nthat transfers from classical tolerance predictors is that the coverage of a\nprediction set based on order statistics, conditional on the calibration set,\nis a random variable stochastically dominating the Beta distribution. We\ndemonstrate the empirical effectiveness of our findings on synthetic and real\ndatasets using a popular split conformal prediction procedure called\nconformalized quantile regression (CQR).\n", "date": "2022-10-26", "categories": []} {"id": "2210.03067", "title": "Few-Shot Calibration of Set Predictors via Meta-Learned\n Cross-Validation-Based Conformal Prediction", "abstract": " Conventional frequentist learning is known to yield poorly calibrated models\nthat fail to reliably quantify the uncertainty of their decisions. Bayesian\nlearning can improve calibration, but formal guarantees apply only under\nrestrictive assumptions about correct model specification. Conformal prediction\n(CP) offers a general framework for the design of set predictors with\ncalibration guarantees that hold regardless of the underlying data generation\nmechanism. However, when training data are limited, CP tends to produce large,\nand hence uninformative, predicted sets. This paper introduces a novel\nmeta-learning solution that aims at reducing the set prediction size. Unlike\nprior work, the proposed meta-learning scheme, referred to as meta-XB, (i)\nbuilds on cross-validation-based CP, rather than the less efficient\nvalidation-based CP; and (ii) preserves formal per-task calibration guarantees,\nrather than less stringent task-marginal guarantees. Finally, meta-XB is\nextended to adaptive non-conformal scores, which are shown empirically to\nfurther enhance marginal per-input calibration.\n", "date": "2022-10-06", "categories": []} {"id": "2209.15145", "title": "Batch Multivalid Conformal Prediction", "abstract": " We develop fast distribution-free conformal prediction algorithms for\nobtaining multivalid coverage on exchangeable data in the batch setting.\nMultivalid coverage guarantees are stronger than marginal coverage guarantees\nin two ways: (1) They hold even conditional on group membership -- that is, the\ntarget coverage level $1-\\alpha$ holds conditionally on membership in each of\nan arbitrary (potentially intersecting) group in a finite collection\n$\\mathcal{G}$ of regions in the feature space. (2) They hold even conditional\non the value of the threshold used to produce the prediction set on a given\nexample. In fact multivalid coverage guarantees hold even when conditioning on\ngroup membership and threshold value simultaneously.\n We give two algorithms: both take as input an arbitrary non-conformity score\nand an arbitrary collection of possibly intersecting groups $\\mathcal{G}$, and\nthen can equip arbitrary black-box predictors with prediction sets. Our first\nalgorithm (BatchGCP) is a direct extension of quantile regression, needs to\nsolve only a single convex minimization problem, and produces an estimator\nwhich has group-conditional guarantees for each group in $\\mathcal{G}$. Our\nsecond algorithm (BatchMVP) is iterative, and gives the full guarantees of\nmultivalid conformal prediction: prediction sets that are valid conditionally\nboth on group membership and non-conformity threshold. We evaluate the\nperformance of both of our algorithms in an extensive set of experiments. Code\nto replicate all of our experiments can be found at\nhttps://github.com/ProgBelarus/BatchMultivalidConformal\n", "date": "2022-09-30", "categories": []} {"id": "2209.03580", "title": "Conformal Methods for Quantifying Uncertainty in Spatiotemporal Data: A\n Survey", "abstract": " Machine learning methods are increasingly widely used in high-risk settings\nsuch as healthcare, transportation, and finance. In these settings, it is\nimportant that a model produces calibrated uncertainty to reflect its own\nconfidence and avoid failures. In this paper we survey recent works on\nuncertainty quantification (UQ) for deep learning, in particular\ndistribution-free Conformal Prediction method for its mathematical properties\nand wide applicability. We will cover the theoretical guarantees of conformal\nmethods, introduce techniques that improve calibration and efficiency for UQ in\nthe context of spatiotemporal data, and discuss the role of UQ in the context\nof safe decision making.\n", "date": "2022-09-08", "categories": []} {"id": "2208.11111", "title": "Integrative conformal p-values for powerful out-of-distribution testing\n with labeled outliers", "abstract": " This paper develops novel conformal methods to test whether a new observation\nwas sampled from the same distribution as a reference set. Blending inductive\nand transductive conformal inference in an innovative way, the described\nmethods can re-weight standard conformal p-values based on dependent side\ninformation from known out-of-distribution data in a principled way, and can\nautomatically take advantage of the most powerful model from any collection of\none-class and binary classifiers. The solution can be implemented either\nthrough sample splitting or via a novel transductive cross-validation+ scheme\nwhich may also be useful in other applications of conformal inference, due to\ntighter guarantees compared to existing cross-validation approaches. After\nstudying false discovery rate control and power within a multiple testing\nframework with several possible outliers, the proposed solution is shown to\noutperform standard conformal p-values through simulations as well as\napplications to image recognition and tabular data.\n", "date": "2022-08-23", "categories": []} {"id": "2207.12274", "title": "MAPIE: an open-source library for distribution-free uncertainty\n quantification", "abstract": " Estimating uncertainties associated with the predictions of Machine Learning\n(ML) models is of crucial importance to assess their robustness and predictive\npower. In this submission, we introduce MAPIE (Model Agnostic Prediction\nInterval Estimator), an open-source Python library that quantifies the\nuncertainties of ML models for single-output regression and multi-class\nclassification tasks. MAPIE implements conformal prediction methods, allowing\nthe user to easily compute uncertainties with strong theoretical guarantees on\nthe marginal coverages and with mild assumptions on the model or on the\nunderlying data distribution. MAPIE is hosted on scikit-learn-contrib and is\nfully \"scikit-learn-compatible\". As such, it accepts any type of regressor or\nclassifier coming with a scikit-learn API. The library is available at:\nhttps://github.com/scikit-learn-contrib/MAPIE/.\n", "date": "2022-07-25", "categories": []} {"id": "2207.05796", "title": "Estimating Test Performance for AI Medical Devices under Distribution\n Shift with Conformal Prediction", "abstract": " Estimating the test performance of software AI-based medical devices under\ndistribution shifts is crucial for evaluating the safety, efficiency, and\nusability prior to clinical deployment. Due to the nature of regulated medical\ndevice software and the difficulty in acquiring large amounts of labeled\nmedical datasets, we consider the task of predicting the test accuracy of an\narbitrary black-box model on an unlabeled target domain without modification to\nthe original training process or any distributional assumptions of the original\nsource data (i.e. we treat the model as a \"black-box\" and only use the\npredicted output responses). We propose a \"black-box\" test estimation technique\nbased on conformal prediction and evaluate it against other methods on three\nmedical imaging datasets (mammography, dermatology, and histopathology) under\nseveral clinically relevant types of distribution shift (institution, hardware\nscanner, atlas, hospital). We hope that by promoting practical and effective\nestimation techniques for black-box models, manufacturers of medical devices\nwill develop more standardized and realistic evaluation procedures to improve\nthe robustness and trustworthiness of clinical AI tools.\n", "date": "2022-07-12", "categories": []} {"id": "2206.11468", "title": "Modular Conformal Calibration", "abstract": " Uncertainty estimates must be calibrated (i.e., accurate) and sharp (i.e.,\ninformative) in order to be useful. This has motivated a variety of methods for\nrecalibration, which use held-out data to turn an uncalibrated model into a\ncalibrated model. However, the applicability of existing methods is limited due\nto their assumption that the original model is also a probabilistic model. We\nintroduce a versatile class of algorithms for recalibration in regression that\nwe call Modular Conformal Calibration (MCC). This framework allows one to\ntransform any regression model into a calibrated probabilistic model. The\nmodular design of MCC allows us to make simple adjustments to existing\nalgorithms that enable well-behaved distribution predictions. We also provide\nfinite-sample calibration guarantees for MCC algorithms. Our framework recovers\nisotonic recalibration, conformal calibration, and conformal interval\nprediction, implying that our theoretical results apply to those methods as\nwell. Finally, we conduct an empirical study of MCC on 17 regression datasets.\nOur results show that new algorithms designed in our framework achieve\nnear-perfect calibration and improve sharpness relative to existing methods.\n", "date": "2022-06-23", "categories": []} {"id": "2206.11810", "title": "Inductive Conformal Prediction: A Straightforward Introduction with\n Examples in Python", "abstract": " Inductive Conformal Prediction (ICP) is a set of distribution-free and model\nagnostic algorithms devised to predict with a user-defined confidence with\ncoverage guarantee. Instead of having point predictions, i.e., a real number in\nthe case of regression or a single class in multi class classification, models\ncalibrated using ICP output an interval or a set of classes, respectively. ICP\ntakes special importance in high-risk settings where we want the true output to\nbelong to the prediction set with high probability. As an example, a\nclassification model might output that given a magnetic resonance image a\npatient has no latent diseases to report. However, this model output was based\non the most likely class, the second most likely class might tell that the\npatient has a 15% chance of brain tumor or other severe disease and therefore\nfurther exams should be conducted. Using ICP is therefore way more informative\nand we believe that should be the standard way of producing forecasts. This\npaper is a hands-on introduction, this means that we will provide examples as\nwe introduce the theory.\n", "date": "2022-06-23", "categories": []} {"id": "2206.12008", "title": "Three Applications of Conformal Prediction for Rating Breast Density in\n Mammography", "abstract": " Breast cancer is the most common cancers and early detection from mammography\nscreening is crucial in improving patient outcomes. Assessing mammographic\nbreast density is clinically important as the denser breasts have higher risk\nand are more likely to occlude tumors. Manual assessment by experts is both\ntime-consuming and subject to inter-rater variability. As such, there has been\nincreased interest in the development of deep learning methods for mammographic\nbreast density assessment. Despite deep learning having demonstrated impressive\nperformance in several prediction tasks for applications in mammography,\nclinical deployment of deep learning systems in still relatively rare;\nhistorically, mammography Computer-Aided Diagnoses (CAD) have over-promised and\nfailed to deliver. This is in part due to the inability to intuitively quantify\nuncertainty of the algorithm for the clinician, which would greatly enhance\nusability. Conformal prediction is well suited to increase reliably and trust\nin deep learning tools but they lack realistic evaluations on medical datasets.\nIn this paper, we present a detailed analysis of three possible applications of\nconformal prediction applied to medical imaging tasks: distribution shift\ncharacterization, prediction quality improvement, and subgroup fairness\nanalysis. Our results show the potential of distribution-free uncertainty\nquantification techniques to enhance trust on AI algorithms and expedite their\ntranslation to usage.\n", "date": "2022-06-23", "categories": []} {"id": "2206.04860", "title": "Conformal Prediction Intervals for Markov Decision Process Trajectories", "abstract": " Before delegating a task to an autonomous system, a human operator may want a\nguarantee about the behavior of the system. This paper extends previous work on\nconformal prediction for functional data and conformalized quantile regression\nto provide conformal prediction intervals over the future behavior of an\nautonomous system executing a fixed control policy on a Markov Decision Process\n(MDP). The prediction intervals are constructed by applying conformal\ncorrections to prediction intervals computed by quantile regression. The\nresulting intervals guarantee that with probability $1-\\delta$ the observed\ntrajectory will lie inside the prediction interval, where the probability is\ncomputed with respect to the starting state distribution and the stochasticity\nof the MDP. The method is illustrated on MDPs for invasive species management\nand StarCraft2 battles.\n", "date": "2022-06-10", "categories": []} {"id": "2206.06584", "title": "Probabilistic Conformal Prediction Using Conditional Random Samples", "abstract": " This paper proposes probabilistic conformal prediction (PCP), a predictive\ninference algorithm that estimates a target variable by a discontinuous\npredictive set. Given inputs, PCP construct the predictive set based on random\nsamples from an estimated generative model. It is efficient and compatible with\neither explicit or implicit conditional generative models. Theoretically, we\nshow that PCP guarantees correct marginal coverage with finite samples.\nEmpirically, we study PCP on a variety of simulated and real datasets. Compared\nto existing methods for conformal inference, PCP provides sharper predictive\nsets.\n", "date": "2022-06-14", "categories": []} {"id": "2206.07851", "title": "Conformal prediction set for time-series", "abstract": " When building either prediction intervals for regression (with real-valued\nresponse) or prediction sets for classification (with categorical responses),\nuncertainty quantification is essential to studying complex machine learning\nmethods. In this paper, we develop Ensemble Regularized Adaptive Prediction Set\n(ERAPS) to construct prediction sets for time-series (with categorical\nresponses), based on the prior work of [Xu and Xie, 2021]. In particular, we\nallow unknown dependencies to exist within features and responses that arrive\nin sequence. Method-wise, ERAPS is a distribution-free and ensemble-based\nframework that is applicable for arbitrary classifiers. Theoretically, we bound\nthe coverage gap without assuming data exchangeability and show asymptotic set\nconvergence. Empirically, we demonstrate valid marginal and conditional\ncoverage by ERAPS, which also tends to yield smaller prediction sets than\ncompeting methods.\n", "date": "2022-06-15", "categories": []} {"id": "2206.01067", "title": "Practical Adversarial Multivalid Conformal Prediction", "abstract": " We give a simple, generic conformal prediction method for sequential\nprediction that achieves target empirical coverage guarantees against\nadversarially chosen data. It is computationally lightweight -- comparable to\nsplit conformal prediction -- but does not require having a held-out validation\nset, and so all data can be used for training models from which to derive a\nconformal score. It gives stronger than marginal coverage guarantees in two\nways. First, it gives threshold calibrated prediction sets that have correct\nempirical coverage even conditional on the threshold used to form the\nprediction set from the conformal score. Second, the user can specify an\narbitrary collection of subsets of the feature space -- possibly intersecting\n-- and the coverage guarantees also hold conditional on membership in each of\nthese subsets. We call our algorithm MVP, short for MultiValid Prediction. We\ngive both theory and an extensive set of empirical evaluations.\n", "date": "2022-06-02", "categories": []} {"id": "2606.22336", "title": "Null-Calibrated Conformal Selection via Target-Membership Scores", "abstract": "Conformal selection aims to identify test candidates whose unknown responses fall in a target region while controlling the false discovery rate. Existing methods often inherit prediction-oriented nonconformity scores, such as residual or clipped residual scores, from conformal prediction. We argue that the natural score for selection is instead the target-membership probability. This score directly addresses the binary event being selected, and any monotone transform of it gives the Neyman--Pearson oracle ranking at a fixed null selection level. This distinction is irrelevant for mean-monotone targets, where conventional scores induce essentially the same ranking, but becomes important for interval-valued, variance-driven, multimodal, or multi-condition targets, where prediction-oriented scores can be misaligned with selection power. We study membership-score-based conformal selection and isolate one conformal calibration route, Null-Calibrated Conformal Selection (NCCS), which ranks test scores against confirmed non-target calibration examples. Under null exchangeability, NCCS yields finite-sample valid null p-values, which can be combined with BY under arbitrary dependence or with BH under standard positive-dependence conditions. Experiments support the score principle: membership scores match conventional scores on mean-monotone targets, substantially improve over mean-score selection on variance-driven targets, and, when calibrated by NCCS, trade power for finite-sample null validity in rare-target regimes where direct empirical-FDP thresholding can be anti-conservative.", "date": "2026-06-21", "categories": []} {"id": "2606.24946", "title": "Conformal Orbit-Valid Trust Horizons for Equivariant World Models", "abstract": "Learned world models are useful only over horizons on which their rollout error remains controlled. We study trust-horizon certification for latent world models with known group symmetries. Given a one-step latent residual and a finite-time expansion estimate, we form a raw horizon curve and calibrate it with a split-conformal multiplicative factor. On the reproducible audit set, the conformal factor is $\\gamma_\\alpha=1.0$: the raw certificate is already conservative under the audit protocol. Across 50 stable audits, we observe zero anti-conservative violations, corresponding to an exact-binomial 95% upper bound of 5.8% on the violation rate. Our main structural result is that exact equivariance transports a calibrated trust-horizon curve over the group orbit: when the environment dynamics, encoder, predictor, action transform, and latent metric satisfy the stated equivariance/invariance conditions, rollout errors and trust horizons are orbit-constant. Empirically, the implemented models exhibit small orbit-transport residuals, with median 1.1% and maximum 4.1% over 14 orbit audits. The certificate is also non-vacuous (median certified-to-measured horizon ratio 0.67). A certificate-level calibration-cost study shows two complementary regimes. On a symmetric 2D substrate, equivariant, plain, and augmented models are all orbit-valid from a single calibration sector -- no separation, because the substrate already makes non-equivariant baselines approximately orbit-robust. A 3D yaw audit shows the other regime: the equivariant model obtains a one-sector safe and non-vacuous orbit-valid certificate, while healthy non-equivariant baselines pay violation, slack, sharpness, or additional-sector cost. The certificate is a conservative, distributional audit rather than a global reachability guarantee, and certificate-guided subgoal spacing is not confirmed in the current 3D CEM-MPC behavior layer.", "date": "2026-06-23", "categories": []} {"id": "2606.24959", "title": "Reliable Conformal Prediction for Ordinal Classification Using the Ranked Probability Score", "abstract": "Ordinal classification (OC) arises in high-stakes domains such as medicine and finance, where uncertainty quantification must account for the severity of ordinal errors. Conformal prediction (CP) provides distribution-free prediction sets with marginal coverage guarantees; however, its practical effectiveness depends critically on the choice of nonconformity function. We introduce a CP method for ordinal classification based on the ranked probability score (RPS), a proper scoring rule defined over cumulative predictive distributions. Although it reflects ordinal risk quite naturally, it has largely been neglected in conformal ordinal prediction (COP). When used as a measure of nonconformity, RPS yields median-centered contiguous prediction sets by construction. The method is model-agnostic, supports both assessed and grouped ordered categorical outcomes, and permits efficient implementation compared to greedy interval selection procedures. Across multiple ordinal image and tabular datasets, RPS-based CP produces contiguous prediction sets and strikes a favorable balance between prediction set width and the magnitude of ordinal miscoverage relative to existing CP methods.", "date": "2026-06-23", "categories": []} {"id": "2606.25371", "title": "Conformal Recovery-Deadline Certificates for Runtime Assurance of Adapting Controllers", "abstract": "Runtime assurance (RTA) protects a safety-critical system by switching from an advanced controller to a verified safe controller when a monitored condition is violated. The standard latching rule, which trips on the first breach of the safe set and then coasts, is correct for a diverging controller but pathological for a capable online-adapting one. Such a controller is unsafe by design during a bounded recovery transient. It must excite the plant to identify the fault before it can correct it, so a latching shield trips on that transient and suppresses a controller that would have recovered. We introduce the conformal recovery-deadline certificate, a split-conformal, distribution-free, finite-sample upper bound on the adapting controller's recovery time that licenses delayed fallback with a coverage guarantee, backstopped by a verified monitor at a hard critical limit. The certified deadline discriminates capable from incapable controllers, keeping the recoverer autonomous while catching the diverger. The construction separates autonomy, governed by statistical coverage, from safety, governed by the verified backstop, as an instance of reliability-asymmetric design. We prove marginal coverage, a weighted extension that restores coverage under a known fault-distribution shift, and group-conditional Mondrian coverage. We demonstrate all three on two unrelated Simplex testbeds: a 6-DOF spacecraft attitude controller and a torque-controlled inverted pendulum. Both show the same suppression pathology and the same cure, making the certificate a domain-general mechanism rather than a single-system trick.", "date": "2026-06-24", "categories": []} {"id": "2606.27001", "title": "Uncertainty quantification via conformal prediction in data assimilation", "abstract": "Quantifying the evolution of uncertainty is critical to both probabilistic forecasting and data assimilation in numerical weather prediction. In this study, we investigate the applicability of conformal prediction (CP), a recent machine learning (ML) method, to quantify uncertainty in a controlled, idealized setting. We use the one dimensional modified shallow water model, designed to mimic the convective process. CP provides a set of possible outcomes with a chosen confidence level. Here, we compare and evaluate the average empirical coverage, the average interval length, miss low, miss high and average interval score loss (AISL) for three variants of CP, namely a) Standard CP, b) Normalized CP and c) Conformalized Quantile Regression. We further compare these CP-based uncertainty estimates with traditional ensemble-based measures such as standard deviation intervals and ensemble spread. In addition, we investigate the integration of CP-derived uncertainty within the data assimilation cycle through CP perturbations. Our results highlight the strengths and limitations of each approach, providing insight into the effectiveness of CP to complement common ensemble-based uncertainty quantification in simplified atmospheric models.", "date": "2026-06-25", "categories": []} {"id": "2606.28598", "title": "Conformal Prediction with Macro-Coverage Guarantees", "abstract": "Prediction sets should have high coverage to be useful, but some coverage notions are more practically relevant than others. In the classification setting, class-conditional coverage requires that the prediction set (i.e., the set of candidate labels for a new test point) must achieve the target accuracy level within each class, which may be challenging to satisfy when many classes are rare and have few calibration points. At the other extreme, marginal coverage requires only that coverage holds on average over the distribution of all classes, which can lead to low-probability labels being essentially ignored. To find a middle ground, recent work has introduced macro-coverage, defined as the unweighted average of class-conditional coverages. Macro-coverage offers a compromise between marginal coverage and class-conditional coverage that is particularly appropriate for long-tailed settings. In this work, we show that label-weighted conformal prediction can be used to produce prediction sets with a finite-sample macro-coverage guarantee, and more generally a guarantee on a family of generalized macro-coverage objectives that aggregate coverage at the level of arbitrary class groupings and take a weighted average. We further characterize the form of the smallest prediction sets satisfying a given generalized macro-coverage objective and propose a corresponding conformal score function. We validate our theoretical results on two large-scale image classification datasets.", "date": "2026-06-26", "categories": []} {"id": "2606.29054", "title": "When Can Conformal Risk Control Certify LLM Outputs? Bounds, Impossibility, and Adaptation for Structured Generation", "abstract": "Large language models (LLMs) deployed for structured generation (NER, JSON extraction, QA, and classification) lack formal reliability guarantees, and standard heuristic abstention policies miss user-specified risk targets by 7.5--12.5%. We characterize when conformal risk control (CRC) can certify structured LLM outputs and when it provably cannot. First, we prove an impossibility result: when the base risk (\\mu > \\alpha), any distribution-free method must abstain on at least ((\\mu-\\alpha)/(1-\\alpha)) examples, yielding a closed-form feasibility test: one can check whether CRC will work before running it. Second, we analyze a certification hierarchy across Hoeffding, empirical Bernstein, and a betting-based e-CRC bound, with strict gains in low-variance/large-sample regimes: the Hoeffding-to-Bernstein step delivers the largest gain (+37% certified configurations), while e-CRC adds value when calibration data is scarce (10% certification at 20% data versus 0% for Hoeffding). Third, we validate adaptive conformal inference (ACI) under cross-dataset shift, reducing risk-target violations from 71% to 21%, with residual failures concentrated exactly where the impossibility bound predicts. Across six open-weight models (3B--72B parameters), eight datasets, four tasks, and six nonconformity scores, hard NER/QA/CLS configurations are uncertifiable at (\\alpha = 0.10); relaxing to (\\alpha = 0.30--0.40) unlocks practical certification (47% NER, 40% QA, 60% CLS). The framework gives a three-step deployment recipe: check feasibility, select the bound and score, then mitigate shift.", "date": "2026-06-27", "categories": []} {"id": "2606.29403", "title": "Self-Organized Conformal Prediction: Reducing Regional Coverage Gaps with Unsupervised Group Discovery", "abstract": "Conformal prediction guarantees marginal coverage, but pooled calibration averages over heterogeneous regions and can mask regional undercoverage in safety-critical subgroups. We introduce Self-Organized Conformal Prediction (SOCP), a calibration scheme that discovers input-space groups with a Self-Organizing Map (SOM) and, at test time, draws a local calibration buffer from the query's best-matching unit (BMU) cell or a fixed grid neighborhood. The same retrieval rule applies to regression and classification tasks across tabular features and image embeddings, leaving the predictor and nonconformity score untouched. SOCP gives exact validity for BMU-cell retrieval and fixed retrieved-set validity for neighborhood buffers; central-cell validity for neighborhood retrieval holds up to a Kolmogorov-Smirnov (KS) bias term. A split-routed extension recovers fixed retrieved-set validity conditional on the routing split. On eight regression and classification benchmarks, SO-SCP reduces the weighted regional coverage gap on $7/8$ datasets (mean paired change $-7.1\\%$) for a mean prediction-set size increase of $6.2\\%$, with negligible overhead on the largest six datasets; SO-CQR yields smaller gains, since quantile regression already absorbs much of the heterogeneity. By learning groups directly from the input geometry, SOCP provides group-local calibration with exact fixed-group guarantees and approximate central-cell guarantees, without supervised partitions or predictor retraining.", "date": "2026-06-28", "categories": []} {"id": "2606.30500", "title": "Doubly Robust Adaptive Conformal Inference for Causal Effects Under Temporal Dependence", "abstract": "We propose doubly robust adaptive conformal inference (DR-ACI), which constructs prediction intervals for doubly robust pseudo-outcomes under temporal dependence.", "date": "2026-06-29", "categories": []} {"id": "2606.31577", "title": "Localized Conformal Prediction for Image Classification with Vision-Language Models", "abstract": "Conformal predictions have attracted significant attention in the field of uncertainty quantification, mainly because of their strong marginal coverage guarantees. Full conditional guarantee is not an attainable goal, a well known fact in conformal predictions literature. As a result, several approaches have tried to approximate this behavior by adapting the conformal sets of test-time samples according to their similarity to calibration examples. Although the latter has gained traction and shown impressive performances for regression problems, its application to image classification remains under-explored. We conduct an extensive benchmarking on natural image classification tasks with vision-language models (VLMs), using our open source implementation of a recent localized conformal prediction algorithm. We show that straightforward usage of the cosine similarity between test-time and calibration visual features, an intuitive choice for VLMs, is not sufficient to improve over the non-local baselines. In response, we propose a simple non-linear transformation of the cosine similarities, which conserves marginal coverage guarantees and achieves statistically significant mean set sizes reduction. Code is available at https://github.com/cfuchs2023/lcp-vlm/.", "date": "2026-06-30", "categories": []} {"id": "2606.31600", "title": "On Optimal Data Splitting for Split Conformal Prediction", "abstract": "Conformal prediction and its variants, including the split conformal prediction, provide a distribution-free framework for uncertainty quantification by constructing prediction intervals or sets with finite-sample coverage guarantees. The statistical efficiency of these intervals depends critically on how the data are split into training and calibration samples. Despite its practical importance, a principled characterization of the training-calibration split that minimizes prediction interval length while maintaining coverage has remained largely unresolved. In this paper, we develop a theoretical framework for optimal data splitting in split conformal prediction. We first analyze the problem in a general setting and derive analytical characterizations of the length-optimal split ratio under both symmetric and asymmetric regimes. We then show how the general results specialize to several commonly used regression settings, including linear regression, nonparametric regression, and neural networks, thereby demonstrating the scope of the framework. We also describe a data-based method for selecting the optimal proportion. Our analysis clarifies how model-related features govern the optimal allocation of samples between training and calibration and provides principled guidance for constructing shorter prediction intervals. Experiments on both synthetic and real-world datasets demonstrate the applicability of the proposed methodology across a variety of practical scenarios.", "date": "2026-06-30", "categories": []} {"id": "2606.31804", "title": "Relational and Sequential Conformal Inference for Energy Time Series over Graphs via Foundation Models", "abstract": "Accurate energy demand forecasting is essential for the reliable operation and planning of modern sustainable energy systems. Spatial-temporal graph neural networks (STGNNs) have recently achieved strong performance in point forecasting by jointly modeling temporal dynamics and relational dependencies across interconnected energy nodes. However, in real-world energy systems, accurate point forecasts alone are insufficient, as operators also require reliable uncertainty estimates to support risk-aware decision-making, grid stability, and operational planning under uncertainty. Conformal prediction provides a principled and model-agnostic framework for uncertainty quantification with statistical coverage guarantees, making it particularly attractive for safety-critical energy applications. However, existing conformal prediction approaches often fail to fully capture the complex spatial-temporal structure of energy systems. To address these limitations, we propose STOIC (Spatial-Temporal Graph Conformal Prediction with In-Context Learning), a novel framework that integrates graph-based forecasting with the zero-shot calibration capabilities of tabular foundation models. STOIC first generates point forecasts using an STGNN and subsequently reformulates spatial-temporal residuals into a tabular representation suitable for in-context learning. Leveraging a tabular foundation model, STOIC calibrates prediction intervals without task-specific retraining, effectively capturing both sequential and relational dependencies. We evaluate STOIC on five diverse benchmarks, including synthetic simulations as well as real-world electricity and district heating networks. Across all datasets, STOIC consistently outperforms existing conformal prediction baselines, delivering more reliable and robust uncertainty estimates for complex graph-structured energy time series.", "date": "2026-06-30", "categories": []} {"id": "2606.31915", "title": "Accelerating Conformal Prediction via Approximate Leave-One-Out", "abstract": "While conformal prediction provides a general framework for uncertainty quantification in predictive inference, its application is often limited by computational cost. Recent methods, including Jackknife+ and Jackknife-minmax, achieve faster computation by trading a slight loss of efficiency relative to full conformal prediction, but still requires computing leave-one-out refits for all observations. In this paper, we further accelerate conformal prediction by incorporating approximate leave-one-out (ALO) estimators, and establish asymptotic coverage and efficiency. While our proof draws on methods developed for analyzing the consistency of ALO cross-validation risk estimators in high-dimensional statistics, it requires adaptations to handle conformal prediction, where leave-$i$-out residuals are needed for predictions at $x_{n+1}$ rather than just at the training covariate $x_i$. Simulation results validate our theoretical findings, showing that the ALO-based methods achieve coverage and efficiency comparable to the exact methods, while significantly reducing the runtime.", "date": "2026-06-30", "categories": []}