中文
相关论文

相关论文: E-Values Expand the Scope of Conformal Prediction

200 篇论文

Conformal prediction is a powerful framework for constructing prediction sets with valid coverage guarantees in multi-class classification. However, existing methods often rely on a single score function, which can limit their efficiency…

机器学习 · 统计学 2025-03-05 Rui Luo , Zhixin Zhou

Typically, machine learning models are trained and evaluated without making any distinction between users (e.g, using traditional hold-out and cross-validation). However, this produces inaccurate performance metrics estimates in multi-user…

机器学习 · 计算机科学 2023-12-11 Enrique Garcia-Ceja , Luciano Garcia-Banuelos , Nicolas Jourdan

While counterfactual fairness of point predictors is well studied, its extension to prediction sets--central to fair decision-making under uncertainty--remains underexplored. On the other hand, conformal prediction (CP) provides efficient,…

机器学习 · 计算机科学 2026-03-13 Ozgur Guldogan , Neeraj Sarna , Yuanyuan Li , Michael Berger

Quantifying uncertainty in automatically generated text is important for letting humans check potential hallucinations and making systems more reliable. Conformal prediction is an attractive framework to provide predictions imbued with…

计算与语言 · 计算机科学 2024-02-02 Dennis Ulmer , Chrysoula Zerva , André F. T. Martins

We develop methods for forming prediction sets in an online setting where the data generating distribution is allowed to vary over time in an unknown fashion. Our framework builds on ideas from conformal inference to provide a general…

统计方法学 · 统计学 2021-12-10 Isaac Gibbs , Emmanuel Candès

Conformal prediction is a popular uncertainty quantification method that augments a base predictor to return sets of predictions with statistically valid coverage guarantees. However, current methods are often computationally expensive and…

机器学习 · 计算机科学 2026-03-05 Laura Lützow , Michael Eichelbeck , Mykel J. Kochenderfer , Matthias Althoff

Conformal prediction methods provide statistically rigorous marginal coverage guarantees for machine learning models, but such guarantees fail to account for algorithmic biases, thereby undermining fairness and trust. This paper introduces…

机器学习 · 计算机科学 2026-05-13 Senrong Xu , Yanke Zhou , Yuhao Tan , Zenan Li , Yuan Yao , Taolue Chen , Feng Xu , Xiaoxing Ma

We introduce the E-measure: a measure-like generalization of the E-value to a class of hypotheses. Unlike classical measures, E-measures are closed under infimums instead of addition. They arise from a compatibility axiom with logical…

统计理论 · 数学 2026-04-23 Nick W. Koning

Conformalized multiple testing offers a model-free way to control predictive uncertainty in decision-making. Existing methods typically use only part of the available data to build score functions tailored to specific settings. We propose a…

统计方法学 · 统计学 2026-05-22 Yuyang Huo , Xiaoyang Wu , Changliang Zou , Haojie Ren

We study e-values for quantifying evidence against exchangeability and general invariance of a random variable under a compact group. We start by characterizing such e-values, and explaining how they nest traditional group invariance tests…

统计理论 · 数学 2026-02-11 Nick W. Koning

We develop E-variables for testing whether two or more data streams come from the same source or not, and more generally, whether the difference between the sources is larger than some minimal effect size. These E-variables lead to exact,…

统计方法学 · 统计学 2022-06-23 Rosanne Turner , Alexander Ly , Peter Grünwald

Compared to p-values, e-values provably guarantee safe, valid inference. If the goal is to test multiple hypotheses simultaneously, one can construct e-values for each individual test and then use the recently developed e-BH procedure to…

统计方法学 · 统计学 2024-12-03 Neil Dey , Ryan Martin , Jonathan P. Williams

Quantifying the data uncertainty in learning tasks is often done by learning a prediction interval or prediction set of the label given the input. Two commonly desired properties for learned prediction sets are \emph{valid coverage} and…

机器学习 · 计算机科学 2022-05-31 Yu Bai , Song Mei , Huan Wang , Yingbo Zhou , Caiming Xiong

Machine-learning techniques are essential in modern collider research, yet their probabilistic outputs often lack calibrated uncertainty estimates and finite-sample guarantees, limiting their direct use in statistical inference and…

高能物理 - 唯象学 · 物理学 2025-12-22 Jack Y. Araz , Michael Spannowsky

Methods for split conformal prediction leverage calibration samples to transform any prediction rule into a set-prediction rule that complies with a target coverage probability. Existing methods provide remarkably strong performance…

机器学习 · 统计学 2025-10-15 Santiago Mazuelas

The problem of individualized prediction can be addressed using variants of conformal prediction, obtaining the intervals to which the actual values of the variables of interest belong. Here we present a method based on detecting the…

统计方法学 · 统计学 2023-04-12 Fernando Delbianco , Fernando Tohmé

Interest has been growing in decision-focused machine learning methods which train models to account for how their predictions are used in downstream optimization problems. Doing so can often improve performance on subsequent decision…

机器学习 · 计算机科学 2025-03-03 Santiago Cortes-Gomez , Carlos Patiño , Yewon Byun , Steven Wu , Eric Horvitz , Bryan Wilder

Conformal prediction is widely adopted in uncertainty quantification, due to its post-hoc, distribution-free, and model-agnostic properties. In the realm of modern deep learning, researchers have proposed Feature Conformal Prediction (FCP),…

机器学习 · 计算机科学 2024-12-03 Zihao Tang , Boyuan Wang , Chuan Wen , Jiaye Teng

This paper presents a conformal prediction method for classification in highly imbalanced and open-set settings, where there are many possible classes and not all may be represented in the data. Existing approaches require a finite, known…

机器学习 · 统计学 2025-10-16 Tianmin Xie , Yanfei Zhou , Ziyi Liang , Stefano Favaro , Matteo Sesia

Conformal prediction provides a powerful framework for constructing distribution-free prediction regions with finite-sample coverage guarantees. While extensively studied in univariate settings, its extension to multi-output problems…

机器学习 · 统计学 2025-02-04 Victor Dheur , Matteo Fontana , Yorick Estievenart , Naomi Desobry , Souhaib Ben Taieb