Skip to main navigation Skip to search Skip to main content

Entropy Reweighted Conformal Classification

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

Abstract

Conformal Prediction (CP) is a powerful framework for constructing prediction sets with guaranteed coverage. However, recent studies have shown that integrating confidence calibration with CP can lead to a degradation in efficiency. In this paper, We propose an adaptive approach that considers the classifier’s uncertainty and employs entropy-based reweighting to enhance the efficiency of prediction sets for conformal classification. Our experimental results demonstrate that this method significantly improves efficiency. © 2024 R. Luo & N. Colombo.
Original languageEnglish
Title of host publicationThe 13th Symposium on Conformal and Probabilistic Prediction with Applications, 9-11 September 2024, Politecnico di Milano, Milano, Italy
PublisherML Research Press
Pages264-276
Publication statusPublished - Sept 2024
Event13th Symposium on Conformal and Probabilistic Prediction with Applications (COPA 2024) - Polytechnic University of Milan, Milan, Italy
Duration: 9 Sept 202411 Sept 2024
https://proceedings.mlr.press/v230/
https://cml.rhul.ac.uk/copa2024/#nav-venue

Publication series

NameProceedings of Machine Learning Research
Volume230
ISSN (Print)2640-3498

Conference

Conference13th Symposium on Conformal and Probabilistic Prediction with Applications (COPA 2024)
PlaceItaly
CityMilan
Period9/09/2411/09/24
Internet address

Research Keywords

  • confidence calibration
  • Conformal prediction
  • entropy reweighting
  • neural networks
  • temperature scaling

Fingerprint

Dive into the research topics of 'Entropy Reweighted Conformal Classification'. Together they form a unique fingerprint.

Cite this