跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2602.04364 · v2

Anytime-Valid Conformal Risk Control

任意时间有效的保形风险控制

首次提交 2026-02-04 17:39
版本更新 2026-11-20 00:16
雷达观测 2026-11-20 12:52
arXiv ↗PDF ↗

摘要

Prediction sets provide a means of quantifying the uncertainty in predictive tasks. Using held out calibration data, conformal prediction and risk control can produce prediction sets that exhibit statistically valid error control in a computationally efficient manner. However, in the standard formulations, the error is only controlled on average over many possible calibration datasets of fixed size. In this paper, we extend the control to remain valid with high probability over a cumulatively growing calibration dataset at any time point. We derive such guarantees using quantile-based arguments and illustrate the applicability of the proposed framework to settings involving distribution shift. We further show that our guarantees are asymptotically tight, in the sense that the risks converge to the target level from below, and that the rate of our correction cannot be improved. Finally, we demonstrate the practical performance of our methods through both simulations and real-world numerical examples.

AI 论文解读

基于摘要 · 非全文审阅

该论文将保形风险控制的错误控制从固定校准集上的平均有效性扩展为在累积增长的校准集上任意时刻以高概率有效。

正在读取账号与研究记录…

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-20 00:16 · 最新观测版本

Anytime-Valid Conformal Risk Control

Prediction sets provide a means of quantifying the uncertainty in predictive tasks. Using held out calibration data, conformal prediction and risk control can produce prediction sets that exhibit statistically valid error control in a computationally efficient manner. However, in the standard formulations, the error is only controlled on average over many possible calibration datasets of fixed size. In this paper, we extend the control to remain valid with high probability over a cumulatively growing calibration dataset at any time point. We derive such guarantees using quantile-based arguments and illustrate the applicability of the proposed framework to settings involving distribution shift. We further show that our guarantees are asymptotically tight, in the sense that the risks converge to the target level from below, and that the rate of our correction cannot be improved. Finally, we demonstrate the practical performance of our methods through both simulations and real-world numerical examples.

查看此版本解读 →查看此版本原文 ↗