GCAD: Anomaly Detection in Multivariate Time Series from the Perspective of Granger Causality

TL;DR

This paper introduces GCAD, a novel anomaly detection method for multivariate time series that leverages interpretable Granger causality to identify changes in causal relationships, improving detection accuracy.

Contribution

The paper presents a new framework that models spatial dependencies through dynamic Granger causality, enhancing interpretability and detection performance over existing correlation-based methods.

Findings

Achieves higher accuracy in anomaly detection on real-world datasets.

Effectively captures causal relationship changes as anomalies.

Provides interpretable causal graphs for better understanding of anomalies.

Abstract

Multivariate time series anomaly detection has numerous real-world applications and is being extensively studied. Modeling pairwise correlations between variables is crucial. Existing methods employ learnable graph structures and graph neural networks to explicitly model the spatial dependencies between variables. However, these methods are primarily based on prediction or reconstruction tasks, which can only learn similarity relationships between sequence embeddings and lack interpretability in how graph structures affect time series evolution. In this paper, we designed a framework that models spatial dependencies using interpretable causal relationships and detects anomalies through changes in causal patterns. Specifically, we propose a method to dynamically discover Granger causality using gradients in nonlinear deep predictors and employ a simple sparsification strategy to obtain a Granger causality graph, detecting anomalies from a causal perspective. Experiments on real-world datasets demonstrate that the proposed model achieves more accurate anomaly detection compared to baseline methods.