NEWMA: a new method for scalable model-free online change-point detection

TL;DR

NEWMA introduces a fast, scalable, model-free online change-point detection method using recursive statistics and random features, suitable for high-dimensional data with limited resources.

Contribution

The paper presents a novel, simple, and efficient change-point detection method that combines EWMA-inspired recursive statistics with random features for high-dimensional data.

Findings

Significantly faster than traditional non-parametric methods.

Maintains high accuracy with limited computational resources.

Effective in high-dimensional streaming data environments.

Abstract

We consider the problem of detecting abrupt changes in the distribution of a multi-dimensional time series, with limited computing power and memory. In this paper, we propose a new, simple method for model-free online change-point detection that relies only on fast and light recursive statistics, inspired by the classical Exponential Weighted Moving Average algorithm (EWMA). The proposed idea is to compute two EWMA statistics on the stream of data with different forgetting factors, and to compare them. By doing so, we show that we implicitly compare recent samples with older ones, without the need to explicitly store them. Additionally, we leverage Random Features (RFs) to efficiently use the Maximum Mean Discrepancy as a distance between distributions, furthermore exploiting recent optical hardware to compute high-dimensional RFs in near constant time. We show that our method is significantly faster than usual non-parametric methods for a given accuracy.