Time Series Change Point Detection with Self-Supervised Contrastive Predictive Coding

TL;DR

This paper introduces TS-CP^2, a self-supervised contrastive learning method for change point detection in time series, outperforming existing approaches by learning representations that distinguish adjacent and separated intervals.

Contribution

The paper presents the first contrastive learning-based approach for time series change point detection, demonstrating significant improvements over state-of-the-art methods.

Findings

Outperforms five state-of-the-art CPD methods across three datasets.

Improves existing methods' F1-score by up to 79.4%.

Effective in diverse real-world time series applications.

Abstract

Change Point Detection (CPD) methods identify the times associated with changes in the trends and properties of time series data in order to describe the underlying behaviour of the system. For instance, detecting the changes and anomalies associated with web service usage, application usage or human behaviour can provide valuable insights for downstream modelling tasks. We propose a novel approach for self-supervised Time Series Change Point detection method based onContrastivePredictive coding (TS-CP^2). TS-CP^2 is the first approach to employ a contrastive learning strategy for CPD by learning an embedded representation that separates pairs of embeddings of time adjacent intervals from pairs of interval embeddings separated across time. Through extensive experiments on three diverse, widely used time series datasets, we demonstrate that our method outperforms five state-of-the-art CPD methods, which include unsupervised and semi-supervisedapproaches. TS-CP^2 is shown to improve the performance of methods that use either handcrafted statistical or temporal features by 79.4% and deep learning-based methods by 17.0% with respect to the F1-score averaged across the three datasets.