Conformal Prediction for Deep Classifier via Label Ranking

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Jianguo, Xi, Huajun, Zhang, Linjun, Yao, Huaxiu, Qiu, Yue, Wei, Hongxin
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914826131865600
author Huang, Jianguo
Xi, Huajun
Zhang, Linjun
Yao, Huaxiu
Qiu, Yue
Wei, Hongxin
author_facet Huang, Jianguo
Xi, Huajun
Zhang, Linjun
Yao, Huaxiu
Qiu, Yue
Wei, Hongxin
contents Conformal prediction is a statistical framework that generates prediction sets containing ground-truth labels with a desired coverage guarantee. The predicted probabilities produced by machine learning models are generally miscalibrated, leading to large prediction sets in conformal prediction. To address this issue, we propose a novel algorithm named $\textit{Sorted Adaptive Prediction Sets}$ (SAPS), which discards all the probability values except for the maximum softmax probability. The key idea behind SAPS is to minimize the dependence of the non-conformity score on the probability values while retaining the uncertainty information. In this manner, SAPS can produce compact prediction sets and communicate instance-wise uncertainty. Extensive experiments validate that SAPS not only lessens the prediction sets but also broadly enhances the conditional coverage rate of prediction sets.
format Preprint
id arxiv_https___arxiv_org_abs_2310_06430
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Conformal Prediction for Deep Classifier via Label Ranking
Huang, Jianguo
Xi, Huajun
Zhang, Linjun
Yao, Huaxiu
Qiu, Yue
Wei, Hongxin
Machine Learning
Computer Vision and Pattern Recognition
Statistics Theory
Conformal prediction is a statistical framework that generates prediction sets containing ground-truth labels with a desired coverage guarantee. The predicted probabilities produced by machine learning models are generally miscalibrated, leading to large prediction sets in conformal prediction. To address this issue, we propose a novel algorithm named $\textit{Sorted Adaptive Prediction Sets}$ (SAPS), which discards all the probability values except for the maximum softmax probability. The key idea behind SAPS is to minimize the dependence of the non-conformity score on the probability values while retaining the uncertainty information. In this manner, SAPS can produce compact prediction sets and communicate instance-wise uncertainty. Extensive experiments validate that SAPS not only lessens the prediction sets but also broadly enhances the conditional coverage rate of prediction sets.
title Conformal Prediction for Deep Classifier via Label Ranking
topic Machine Learning
Computer Vision and Pattern Recognition
Statistics Theory
url https://arxiv.org/abs/2310.06430