PAC Prediction Sets Under Label Shift

• URL: http://arxiv.org/abs/2310.12964v1
• Date: Thu, 19 Oct 2023 17:57:57 GMT
• Title: PAC Prediction Sets Under Label Shift
• Authors: Wenwen Si, Sangdon Park, Insup Lee, Edgar Dobriban and Osbert Bastani
• Abstract summary: Prediction sets capture uncertainty by predicting sets of labels rather than individual labels. We propose a novel algorithm for constructing prediction sets with PAC guarantees in the label shift setting. We evaluate our approach on five datasets.
• Score: 52.30074177997787
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Prediction sets capture uncertainty by predicting sets of labels rather than individual labels, enabling downstream decisions to conservatively account for all plausible outcomes. Conformal inference algorithms construct prediction sets guaranteed to contain the true label with high probability. These guarantees fail to hold in the face of distribution shift, which is precisely when reliable uncertainty quantification can be most useful. We propose a novel algorithm for constructing prediction sets with PAC guarantees in the label shift setting. This method estimates the predicted probabilities of the classes in a target domain, as well as the confusion matrix, then propagates uncertainty in these estimates through a Gaussian elimination algorithm to compute confidence intervals for importance weights. Finally, it uses these intervals to construct prediction sets. We evaluate our approach on five datasets: the CIFAR-10, ChestX-Ray and Entity-13 image datasets, the tabular CDC Heart dataset, and the AGNews text dataset. Our algorithm satisfies the PAC guarantee while producing smaller, more informative, prediction sets compared to several baselines.

Related papers

• Provably Reliable Conformal Prediction Sets in the Presence of Data Poisoning [53.42244686183879]
  Conformal prediction provides model-agnostic and distribution-free uncertainty quantification. Yet, conformal prediction is not reliable under poisoning attacks where adversaries manipulate both training and calibration data. We propose reliable prediction sets (RPS): the first efficient method for constructing conformal prediction sets with provable reliability guarantees under poisoning.
  arXiv  Detail & Related papers  (2024-10-13T15:37:11Z)
• Stochastic Online Conformal Prediction with Semi-Bandit Feedback [29.334511328067777]
  We consider the online learning setting, where examples arrive over time, and the goal is to construct prediction sets dynamically. We propose a novel conformal prediction algorithm targeted at this setting, and prove that it obtains sublinear regret compared to the optimal conformal predictor.
  arXiv  Detail & Related papers  (2024-05-22T00:42:49Z)
• A Conformal Prediction Score that is Robust to Label Noise [13.22445242068721]
  We introduce a conformal score that is robust to label noise. The noise-free conformal score is estimated using the noisy labeled data and the noise level. We show that our method outperforms current methods by a large margin, in terms of the average size of the prediction set.
  arXiv  Detail & Related papers  (2024-05-04T12:22:02Z)
• Conformal Prediction for Deep Classifier via Label Ranking [29.784336674173616]
  Conformal prediction is a statistical framework that generates prediction sets with a desired coverage guarantee. We propose a novel algorithm named $textitSorted Adaptive Prediction Sets$ (SAPS) SAPS discards all the probability values except for the maximum softmax probability.
  arXiv  Detail & Related papers  (2023-10-10T08:54:14Z)
• Practical Adversarial Multivalid Conformal Prediction [27.179891682629183]
  We give a generic conformal prediction method for sequential prediction. It achieves target empirical coverage guarantees against adversarially chosen data. It is computationally lightweight -- comparable to split conformal prediction.
  arXiv  Detail & Related papers  (2022-06-02T14:33:00Z)
• Towards PAC Multi-Object Detection and Tracking [26.19794470266982]
  We consider a strategy known as conformal prediction, which predicts sets of labels instead of a single label. We propose multi-object detection and tracking algorithms that come with probably approximately correct (PAC) guarantees. We empirically demonstrate that our method can detect and track objects with PAC guarantees on the COCO and MOT-17 datasets.
  arXiv  Detail & Related papers  (2022-04-15T14:33:42Z)
• Efficient and Differentiable Conformal Prediction with General Function Classes [96.74055810115456]
  We propose a generalization of conformal prediction to multiple learnable parameters. We show that it achieves approximate valid population coverage and near-optimal efficiency within class. Experiments show that our algorithm is able to learn valid prediction sets and improve the efficiency significantly.
  arXiv  Detail & Related papers  (2022-02-22T18:37:23Z)
• Distribution-free uncertainty quantification for classification under label shift [105.27463615756733]
  We focus on uncertainty quantification (UQ) for classification problems via two avenues. We first argue that label shift hurts UQ, by showing degradation in coverage and calibration. We examine these techniques theoretically in a distribution-free framework and demonstrate their excellent practical performance.
  arXiv  Detail & Related papers  (2021-03-04T20:51:03Z)
• Distribution-Free, Risk-Controlling Prediction Sets [112.9186453405701]
  We show how to generate set-valued predictions from a black-box predictor that control the expected loss on future test points at a user-specified level. Our approach provides explicit finite-sample guarantees for any dataset by using a holdout set to calibrate the size of the prediction sets.
  arXiv  Detail & Related papers  (2021-01-07T18:59:33Z)
• Uncertainty Sets for Image Classifiers using Conformal Prediction [112.54626392838163]
  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.
  arXiv  Detail & Related papers  (2020-09-29T17:58:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.