Time Series Change Point Detection with Self-Supervised Contrastive Predictive Coding
This work provides a substantial improvement in change point detection accuracy for practitioners working with time series data, particularly in web service usage and human behavior analysis.
This paper addresses the problem of Time Series Change Point Detection (CPD) by introducing TS-CP^2, a self-supervised method based on Contrastive Predictive Coding. The method significantly outperforms existing state-of-the-art approaches, improving F1-score by 79.4% over methods using handcrafted features and by 17.0% over deep learning-based methods.
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.