Entropy Reweighted Conformal Classification

Rui Luo, Nicolo Colombo
Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications, PMLR 230:264-276, 2024.

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v230-luo24a, title = {Entropy Reweighted Conformal Classification}, author = {Luo, Rui and Colombo, Nicolo}, booktitle = {Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications}, pages = {264--276}, year = {2024}, editor = {Vantini, Simone and Fontana, Matteo and Solari, Aldo and Boström, Henrik and Carlsson, Lars}, volume = {230}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v230/main/assets/luo24a/luo24a.pdf}, url = {https://proceedings.mlr.press/v230/luo24a.html}, 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.} }
Endnote
%0 Conference Paper %T Entropy Reweighted Conformal Classification %A Rui Luo %A Nicolo Colombo %B Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications %C Proceedings of Machine Learning Research %D 2024 %E Simone Vantini %E Matteo Fontana %E Aldo Solari %E Henrik Boström %E Lars Carlsson %F pmlr-v230-luo24a %I PMLR %P 264--276 %U https://proceedings.mlr.press/v230/luo24a.html %V 230 %X 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.
APA
Luo, R. & Colombo, N.. (2024). Entropy Reweighted Conformal Classification. Proceedings of the Thirteenth Symposium on Conformal and Probabilistic Prediction with Applications, in Proceedings of Machine Learning Research 230:264-276 Available from https://proceedings.mlr.press/v230/luo24a.html.

Related Material