Conformal Prediction for Ensembles: Improving Efficiency via Score-Based Aggregation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rivera, Eduardo Ochoa, Patel, Yash, Tewari, Ambuj
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909620478410752
author Rivera, Eduardo Ochoa
Patel, Yash
Tewari, Ambuj
author_facet Rivera, Eduardo Ochoa
Patel, Yash
Tewari, Ambuj
contents Distribution-free uncertainty estimation for ensemble methods is increasingly desirable due to the widening deployment of multi-modal black-box predictive models. Conformal prediction is one approach that avoids such distributional assumptions. Methods for conformal aggregation have in turn been proposed for ensembled prediction, where the prediction regions of individual models are merged as to retain coverage guarantees while minimizing conservatism. Merging the prediction regions directly, however, sacrifices structures present in the conformal scores that can further reduce conservatism. We, therefore, propose a novel framework that extends the standard scalar formulation of a score function to a multivariate score that produces more efficient prediction regions. We then demonstrate that such a framework can be efficiently leveraged in both classification and predict-then-optimize regression settings downstream and empirically show the advantage over alternate conformal aggregation methods.
format Preprint
id arxiv_https___arxiv_org_abs_2405_16246
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Conformal Prediction for Ensembles: Improving Efficiency via Score-Based Aggregation
Rivera, Eduardo Ochoa
Patel, Yash
Tewari, Ambuj
Methodology
Machine Learning
Distribution-free uncertainty estimation for ensemble methods is increasingly desirable due to the widening deployment of multi-modal black-box predictive models. Conformal prediction is one approach that avoids such distributional assumptions. Methods for conformal aggregation have in turn been proposed for ensembled prediction, where the prediction regions of individual models are merged as to retain coverage guarantees while minimizing conservatism. Merging the prediction regions directly, however, sacrifices structures present in the conformal scores that can further reduce conservatism. We, therefore, propose a novel framework that extends the standard scalar formulation of a score function to a multivariate score that produces more efficient prediction regions. We then demonstrate that such a framework can be efficiently leveraged in both classification and predict-then-optimize regression settings downstream and empirically show the advantage over alternate conformal aggregation methods.
title Conformal Prediction for Ensembles: Improving Efficiency via Score-Based Aggregation
topic Methodology
Machine Learning
url https://arxiv.org/abs/2405.16246