Pith. sign in

REVIEW 17 cited by

Uncertainty Sets for Image Classifiers using Conformal Prediction

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2009.14193 v5 pith:GEZBR7V7 submitted 2020-09-29 cs.CV math.STstat.MLstat.TH

Anastasios Angelopoulos , Stephen Bates , Jitendra Malik , Michael I. Jordan This is my paper · ORCID
classification cs.CVmath.STstat.MLstat.TH
keywords plattscalingalgorithmclassifiersexistingpredictivesetsuncertainty
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Convolutional image classifiers can achieve high predictive accuracy, but quantifying their uncertainty remains an unresolved challenge, hindering their deployment in consequential settings. Existing uncertainty quantification techniques, such as Platt scaling, attempt to calibrate the network's probability estimates, but they do not have formal guarantees. We present an algorithm that modifies any classifier to output a predictive set containing the true label with a user-specified probability, such as 90%. The algorithm is simple and fast like Platt scaling, but provides a formal finite-sample coverage guarantee for every model and dataset. Our method modifies an existing conformal prediction algorithm to give more stable predictive sets by regularizing the small scores of unlikely classes after Platt scaling. In experiments on both Imagenet and Imagenet-V2 with ResNet-152 and other classifiers, our scheme outperforms existing approaches, achieving coverage with sets that are often factors of 5 to 10 smaller than a stand-alone Platt scaling baseline.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 17 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Improving Backward Conformal Prediction via Non-Conformity Score Transformation

    stat.ML 2026-02 reject novelty 7.0 of 10

    ST-BCP tightens the coverage bound in Backward Conformal Prediction by applying a computable data-dependent transformation to nonconformity scores, reducing the average gap from 4.20% to 1.12% on benchmarks while prov...

  2. Shadows of Uniform Hypergraphs under a Minimum Degree Condition

    math.CO 2026-05 unverdicted novelty 6.0 of 10

    Under a minimum-degree condition, for large enough ground sets there is an extremal hypergraph containing an isolated K^k_{t+1}, with a smaller size threshold than prior work for triples.

  3. SpeedCP: Fast Kernel-based Conditional Conformal Prediction

    stat.ME 2025-09 conditional novelty 6.0 of 10

    SpeedCP traces the regularization and score solution paths of RKHS quantile regression, making RKHS-based conditional conformal prediction fast and adaptable to low-rank latent embeddings.

  4. FedCF: Fair Federated Conformal Prediction

    cs.LG 2025-09 conditional novelty 6.0 of 10

    FedCF brings Conformal Fairness into federated learning by decomposing the group coverage gap into client-computable terms, then validating the framework across tabular, graph, and image datasets.

  5. ConformalSAM: Unlocking the Potential of Foundational Segmentation Models in Semi-Supervised Semantic Segmentation with Conformal Prediction

    cs.CV 2025-07 conditional novelty 6.0 of 10

    ConformalSAM calibrates SEEM's pseudo-labels with conformal prediction on the labeled set and later switches to self-training, improving semi-supervised semantic segmentation on VOC and ADE20K.

  6. QUTCC: Quantile Uncertainty Training and Conformal Calibration for Imaging Inverse Problems

    eess.IV 2025-07 conditional novelty 6.0 of 10

    QUTCC combines simultaneous quantile regression with conformal calibration of the quantile conditioning inputs to produce spatially adaptive, marginally calibrated uncertainty intervals for imaging inverse problems.

  7. Conformal Prediction Beyond the Seen: A Missing Mass Perspective for Uncertainty Quantification in Generative Models

    cs.LG 2025-06 reject novelty 6.0 of 10

    CPQ builds conformal prediction sets for black-box LLMs by stopping queries when the estimated missing-mass derivative is small and thresholding a Good-Turing based score, with a fallback EE label for unseen correct answers.

  8. Test-time augmentation improves efficiency in conformal prediction

    cs.LG 2025-05 conditional novelty 6.0 of 10

    Applying learned test-time augmentation before conformal scoring reduces prediction set sizes by 10-14% with no loss of nominal coverage.

  9. Optimal Conformal Prediction under Epistemic Uncertainty

    stat.ML 2025-05 conditional novelty 6.0 of 10

    Bernoulli Prediction Sets provably produce smallest conformal prediction sets with conditional coverage when the model outputs valid credal sets, generalizing APS.

  10. Domain-Shift-Aware Conformal Prediction for Large Language Models

    stat.ML 2025-10 reject novelty 5.0 of 10

    DS-CP reweights calibration scores via embedding-based density ratios to improve conformal coverage under domain shift, but its stated guarantee requires a weight condition that the default configuration violates.

  11. Explainable Evidential Clustering

    cs.LG 2025-07 conditional novelty 5.0 of 10

    This paper proposes IEMM, a decision-tree explainer for evidential clustering that minimizes a utility-weighted 'evidential mistakeness' cost.

  12. Uncertainty-Aware Prediction of Parkinson's Disease Medication Needs: A Two-Stage Conformal Prediction Approach

    cs.LG 2025-08 unverdicted novelty 4.0 of 10

    A two-stage conformal model predicts whether Parkinson's patients need medication changes and bounds future levodopa-equivalent dose adjustments with calibrated intervals.

  13. Non-exchangeable Conformal Prediction for Temporal Graph Neural Networks

    cs.LG 2025-07 reject novelty 4.0 of 10

    NCPNet applies non-exchangeable conformal prediction to temporal graphs by diffusing non-conformity scores over graph and time neighbors and learning weighted quantiles to reduce prediction set size.

  14. Residual Reweighted Conformal Prediction for Graph Neural Networks

    cs.LG 2025-06 reject novelty 4.0 of 10

    RR-GNN uses a residual-predicting GNN and graph clustering to produce tighter conformal prediction intervals for GNN outputs while preserving marginal coverage guarantees.

  15. WQLCP: Weighted Adaptive Conformal Prediction for Robust Uncertainty Quantification Under Distribution Shifts

    cs.LG 2025-05 reject novelty 4.0 of 10

    WQLCP weights calibration samples by VAE reconstruction losses and scales test scores by a test-loss quantile to improve conformal prediction under shifts, but the algorithm is ill-defined and the empirical support is weak.

  16. Pitfalls of Conformal Predictions for Medical Image Classification

    cs.LG 2025-06 conditional novelty 3.0 of 10

    Conformal prediction's marginal coverage guarantee does not extend to distribution shifts, subgroups, or selective classification, and it has limited practical value in few-class medical tasks.

  17. Domain Adaptive Skin Lesion Classification via Conformal Ensemble of Vision Transformers

    cs.CV 2025-05 reject novelty 3.0 of 10

    An ensemble of three vision transformers with conformal prediction achieves 90.38% coverage on skin-lesion classification, but the reported improvement over single models is not evaluated with a fixed error rate.

Pith tools