Subspace Change-Point Detection via Low-Rank Matrix Factorisation

TL;DR

This paper introduces a method for detecting multiple change-points in high-dimensional multivariate time series by leveraging low-rank matrix factorisation to identify shifts in the underlying subspace structure.

Contribution

It proposes a novel change-point detection technique that effectively captures subspace changes in high-dimensional data using low-rank matrix factorisation.

Findings

Effective detection of multiple subspace change-points.

Outperforms several state-of-the-art methods.

Validated on synthetic and real datasets.

Abstract

Multivariate time series can often have a large number of dimensions, whether it is due to the vast amount of collected features or due to how the data sources are processed. Frequently, the main structure of the high-dimensional time series can be well represented by a lower dimensional subspace. As vast quantities of data are being collected over long periods of time, it is reasonable to assume that the underlying subspace structure would change over time. In this work, we propose a change-point detection method based on low-rank matrix factorisation that can detect multiple changes in the underlying subspace of a multivariate time series. Experimental results on both synthetic and real data sets demonstrate the effectiveness of our approach and its advantages against various state-of-the-art methods.