Evidential Uncertainty Sets in Deep Classifiers Using Conformal Prediction

Hamed Karimi, Reza Samavi
6/16/2024
cs.LGcs.AIcs.CV

No official code found. Best implementation: tailabTMU/ECP (0 stars, last commit 2024-07-16).

Abstract

In this paper, we propose Evidential Conformal Prediction (ECP) method for image classifiers to generate the conformal prediction sets. Our method is designed based on a non-conformity score function that has its roots in Evidential Deep Learning (EDL) as a method of quantifying model (epistemic) uncertainty in DNN classifiers. We use evidence that are derived from the logit values of target labels to compute the components of our non-conformity score function: the heuristic notion of uncertainty in CP, uncertainty surprisal, and expected utility. Our extensive experimental evaluation demonstrates that ECP outperforms three state-of-the-art methods for generating CP sets, in terms of their set sizes and adaptivity while maintaining the coverage of true labels.

View on arXivView PDF

Code Implementations(1)

tailabTMU/ECPCommunity100%
00PythonJun 25, 20242 years ago

Similar Papers

Quantifying Deep Learning Model Uncertainty in Conformal Prediction

89% match2023

Improving Uncertainty Quantification of Deep Classifiers via Neighborhood Conformal Prediction: Novel Algorithm and Theoretical Analysis

89% match1 repo102023

Evaluating the Utility of Conformal Prediction Sets for AI-Advised Image Labeling

88% match1 repo62024

Trustworthy Classification through Rank-Based Conformal Prediction Sets

86% match2024

Class Adaptive Conformal Training

86% match2026

Knowledge from Uncertainty in Evidential Deep Learning

85% match2023

Cite this paper

@article{karimi2024evidential,
  title  = {Evidential Uncertainty Sets in Deep Classifiers Using Conformal Prediction},
  author = {Hamed Karimi and Reza Samavi},
  year   = {2024},
  eprint = {2406.10787},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.10787v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.10787.svg)](https://codeofpaper.com/paper/2406.10787)

Discussion