Saved in:
Bibliographic Details
Main Authors: Barber, Rina Foygel, Tibshirani, Ryan J.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.02292
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918258267914240
author Barber, Rina Foygel
Tibshirani, Ryan J.
author_facet Barber, Rina Foygel
Tibshirani, Ryan J.
contents This paper presents a unified framework for understanding the methodology and theory behind several different methods in the conformal prediction literature, which includes standard conformal prediction (CP), weighted conformal prediction (WCP), nonexchangeable conformal prediction (NexCP), and randomly-localized conformal prediction (RLCP), among others. At the crux of our framework is the idea that conformal methods are based on revealing partial information about the data at hand, and positing a conditional distribution for the data given the partial information. Different methods arise from different choices of partial information, and of the corresponding (approximate) conditional distribution. In addition to recovering and unifying existing results, our framework leads to both new theoretical guarantees for existing methods, and new extensions of the conformal methodology.
format Preprint
id arxiv_https___arxiv_org_abs_2504_02292
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unifying Different Theories of Conformal Prediction
Barber, Rina Foygel
Tibshirani, Ryan J.
Statistics Theory
This paper presents a unified framework for understanding the methodology and theory behind several different methods in the conformal prediction literature, which includes standard conformal prediction (CP), weighted conformal prediction (WCP), nonexchangeable conformal prediction (NexCP), and randomly-localized conformal prediction (RLCP), among others. At the crux of our framework is the idea that conformal methods are based on revealing partial information about the data at hand, and positing a conditional distribution for the data given the partial information. Different methods arise from different choices of partial information, and of the corresponding (approximate) conditional distribution. In addition to recovering and unifying existing results, our framework leads to both new theoretical guarantees for existing methods, and new extensions of the conformal methodology.
title Unifying Different Theories of Conformal Prediction
topic Statistics Theory
url https://arxiv.org/abs/2504.02292