ABACUS: Unsupervised Multivariate Change Detection via Bayesian Source Separation

TL;DR

ABACUS is a Bayesian source separation method that unsupervisedly detects multivariate change points such as outliers and level shifts, effectively handling correlated changes and noise in high-dimensional data.

Contribution

It introduces ABACUS, a novel unsupervised Bayesian approach with multi-level sparsity for multivariate change detection and latent signal recovery.

Findings

Outperforms state-of-the-art methods in simulations

Effective in real genomic and electricity data

Handles correlated change points and noise robustly

Abstract

Change detection involves segmenting sequential data such that observations in the same segment share some desired properties. Multivariate change detection continues to be a challenging problem due to the variety of ways change points can be correlated across channels and the potentially poor signal-to-noise ratio on individual channels. In this paper, we are interested in locating additive outliers (AO) and level shifts (LS) in the unsupervised setting. We propose ABACUS, Automatic BAyesian Changepoints Under Sparsity, a Bayesian source separation technique to recover latent signals while also detecting changes in model parameters. Multi-level sparsity achieves both dimension reduction and modeling of signal changes. We show ABACUS has competitive or superior performance in simulation studies against state-of-the-art change detection methods and established latent variable models. We also illustrate ABACUS on two real application, modeling genomic profiles and analyzing household electricity consumption.