Conformal prediction using conditional histograms

作     者：Sesia, Matteo Romano, Yaniv 

作者机构：Department of Data Sciences and Operations University of Southern California United States Departments of Electrical and of Computer Engineering and of Computer Science Technion Israel 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2021年

核心收录：

主　　题：Sampling 

摘      要：This paper develops a conformal method to compute prediction intervals for non-parametric regression that can automatically adapt to skewed data. Leveraging black-box machine learning algorithms to estimate the conditional distribution of the outcome using histograms, it translates their output into the shortest prediction intervals with approximate conditional coverage. The resulting prediction intervals provably have marginal coverage in finite samples, while asymptotically achieving conditional coverage and optimal length if the black-box model is consistent. Numerical experiments with simulated and real data demonstrate improved performance compared to state-of-the-art alternatives, including conformalized quantile regression and other distributional conformal prediction approaches. Copyright © 2021, The Authors. All rights reserved.