On Rank Energy Statistics via Optimal Transport: Continuity, Convergence, and Change Point Detection

TL;DR

This paper explores the use of optimal transport-based rank energy statistics for nonparametric change point detection, highlighting the advantages of soft rank energy in terms of convergence, robustness, and reduced false alarms.

Contribution

It introduces the soft rank energy statistic with proven fast convergence and robustness, and analyzes its advantages over traditional rank energy in change point detection.

Findings

Soft rank energy has fast statistical convergence and robust continuity.

Traditional rank energy suffers from curse of dimensionality and false alarms.

The proposed method outperforms other optimal transport-based methods and MMD in experiments.

Abstract

This paper considers the use of recently proposed optimal transport-based multivariate test statistics, namely rank energy and its variant the soft rank energy derived from entropically regularized optimal transport, for the unsupervised nonparametric change point detection (CPD) problem. We show that the soft rank energy enjoys both fast rates of statistical convergence and robust continuity properties which lead to strong performance on real datasets. Our theoretical analyses remove the need for resampling and out-of-sample extensions previously required to obtain such rates. In contrast the rank energy suffers from the curse of dimensionality in statistical estimation and moreover can signal a change point from arbitrarily small perturbations, which leads to a high rate of false alarms in CPD. Additionally, under mild regularity conditions, we quantify the discrepancy between soft rank energy and rank energy in terms of the regularization parameter. Finally, we show our approach performs favorably in numerical experiments compared to several other optimal transport-based methods as well as maximum mean discrepancy.