Correlation-aware Unsupervised Change-point Detection via Graph Neural Networks
This work addresses a key limitation in change-point detection for multivariate time series, offering a more realistic approach for applications like anomaly detection, though it is incremental in combining graph neural networks with existing frameworks.
The paper tackled the problem of detecting abrupt changes in multivariate time series by explicitly modeling dynamic correlation structures, which existing methods ignore or assume static, and demonstrated improved performance over strong baselines with the ability to classify change-point types.
Change-point detection (CPD) aims to detect abrupt changes over time series data. Intuitively, effective CPD over multivariate time series should require explicit modeling of the dependencies across input variables. However, existing CPD methods either ignore the dependency structures entirely or rely on the (unrealistic) assumption that the correlation structures are static over time. In this paper, we propose a Correlation-aware Dynamics Model for CPD, which explicitly models the correlation structure and dynamics of variables by incorporating graph neural networks into an encoder-decoder framework. Extensive experiments on synthetic and real-world datasets demonstrate the advantageous performance of the proposed model on CPD tasks over strong baselines, as well as its ability to classify the change-points as correlation changes or independent changes. Keywords: Multivariate Time Series, Change-point Detection, Graph Neural Networks