Validity and efficiency of the conformal CUSUM procedure

Vladimir Vovk; Ilia Nouretdinov; Alex Gammerman

Validity and efficiency of the conformal CUSUM procedure

Vladimir Vovk, Ilia Nouretdinov, Alex Gammerman

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

Abstract

In this paper we study the validity and efficiency of a conformal version of the CUSUM procedure for change detection both experimentally and theoretically. Unlike the standard CUSUM procedure, its conformal version tests repeatedly a massive null hypothesis, that of the data being IID. We establish a novel property of validity for the conformal CUSUM procedure and for the first time establish a property of efficiency.

Original language	English
Title of host publication	Proceedings of Machine Learning Research (COPA 2025)
Editors	Khuong An Nguyen, Zhiyuan Luo, Harris Papadopoulos, Tuwe Lofstrom, Lars Carlsson, Hentrik Bostrom
Pages	576-594
Volume	266
Publication status	Published - 15 Aug 2025

Access to Document

https://proceedings.mlr.press/v266/vovk25b.html

Cite this

@inbook{e1ec11eee30a4389986b88bcf8fe20a2,

title = "Validity and efficiency of the conformal CUSUM procedure",

abstract = "In this paper we study the validity and efficiency of a conformal version of the CUSUM procedure for change detection both experimentally and theoretically. Unlike the standard CUSUM procedure, its conformal version tests repeatedly a massive null hypothesis, that of the data being IID. We establish a novel property of validity for the conformal CUSUM procedure and for the first time establish a property of efficiency. ",

author = "Vladimir Vovk and Ilia Nouretdinov and Alex Gammerman",

year = "2025",

month = aug,

day = "15",

language = "English",

volume = "266",

pages = "576--594",

editor = "\{An Nguyen\}, Khuong and Zhiyuan Luo and Harris Papadopoulos and Tuwe Lofstrom and Lars Carlsson and Hentrik Bostrom",

booktitle = "Proceedings of Machine Learning Research (COPA 2025)",

}

Validity and efficiency of the conformal CUSUM procedure. / Vovk, Vladimir ; Nouretdinov, Ilia ; Gammerman, Alex.
Proceedings of Machine Learning Research (COPA 2025). ed. / Khuong An Nguyen; Zhiyuan Luo; Harris Papadopoulos; Tuwe Lofstrom; Lars Carlsson; Hentrik Bostrom. Vol. 266 2025. p. 576-594.

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

TY - CHAP

T1 - Validity and efficiency of the conformal CUSUM procedure

AU - Vovk, Vladimir

AU - Nouretdinov, Ilia

AU - Gammerman, Alex

PY - 2025/8/15

Y1 - 2025/8/15

N2 - In this paper we study the validity and efficiency of a conformal version of the CUSUM procedure for change detection both experimentally and theoretically. Unlike the standard CUSUM procedure, its conformal version tests repeatedly a massive null hypothesis, that of the data being IID. We establish a novel property of validity for the conformal CUSUM procedure and for the first time establish a property of efficiency.

AB - In this paper we study the validity and efficiency of a conformal version of the CUSUM procedure for change detection both experimentally and theoretically. Unlike the standard CUSUM procedure, its conformal version tests repeatedly a massive null hypothesis, that of the data being IID. We establish a novel property of validity for the conformal CUSUM procedure and for the first time establish a property of efficiency.

M3 - Chapter (peer-reviewed)

VL - 266

SP - 576

EP - 594

BT - Proceedings of Machine Learning Research (COPA 2025)

A2 - An Nguyen, Khuong

A2 - Luo, Zhiyuan

A2 - Papadopoulos, Harris

A2 - Lofstrom, Tuwe

A2 - Carlsson, Lars

A2 - Bostrom, Hentrik

ER -