Optimal Online Change Detection via Random Fourier Features

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kalinke, Florian, Gavioli-Akilagun, Shakeel
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917049752616960
author Kalinke, Florian
Gavioli-Akilagun, Shakeel
author_facet Kalinke, Florian
Gavioli-Akilagun, Shakeel
contents This article studies the problem of online non-parametric change point detection in multivariate data streams. We approach the problem through the lens of kernel-based two-sample testing and introduce a sequential testing procedure based on random Fourier features, running with logarithmic time complexity per observation and with overall logarithmic space complexity. The algorithm has two advantages compared to the state of the art. First, our approach is genuinely online, and no access to training data known to be from the pre-change distribution is necessary. Second, the algorithm does not require the user to specify a window parameter over which local tests are to be calculated. We prove strong theoretical guarantees on the algorithm's performance, including information-theoretic bounds demonstrating that the detection delay is optimal in the minimax sense. Numerical studies on real and synthetic data show that our algorithm is competitive with respect to the state of the art.
format Preprint
id arxiv_https___arxiv_org_abs_2505_17789
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Optimal Online Change Detection via Random Fourier Features
Kalinke, Florian
Gavioli-Akilagun, Shakeel
Machine Learning
68W27 (Primary) 62G10, 46E22 (Secondary)
G.3; I.2.6
This article studies the problem of online non-parametric change point detection in multivariate data streams. We approach the problem through the lens of kernel-based two-sample testing and introduce a sequential testing procedure based on random Fourier features, running with logarithmic time complexity per observation and with overall logarithmic space complexity. The algorithm has two advantages compared to the state of the art. First, our approach is genuinely online, and no access to training data known to be from the pre-change distribution is necessary. Second, the algorithm does not require the user to specify a window parameter over which local tests are to be calculated. We prove strong theoretical guarantees on the algorithm's performance, including information-theoretic bounds demonstrating that the detection delay is optimal in the minimax sense. Numerical studies on real and synthetic data show that our algorithm is competitive with respect to the state of the art.
title Optimal Online Change Detection via Random Fourier Features
topic Machine Learning
68W27 (Primary) 62G10, 46E22 (Secondary)
G.3; I.2.6
url https://arxiv.org/abs/2505.17789