Causality-Aware Contrastive Learning for Robust Multivariate Time-Series Anomaly Detection

2025-06-04

AI Summary

Existing multivariate time series anomaly detection (MTSAD) methods suffer from inadequate causal modeling and limited robustness to noise and non-stationarity. Method: This paper introduces the first causal-aware contrastive learning paradigm for MTSAD, integrating causal inference into contrastive learning. It discovers causal structures using the PC algorithm and a Granger-causality variant; generates positive samples via causal-preserving augmentations and negative samples via causal perturbations; and achieves causal-consistent separation of normal and anomalous patterns in the latent space. It further proposes a similarity-filtered one-class contrastive loss and a causal-structure-guided anomaly discrimination mechanism. Results: Evaluated on five real-world and two synthetic datasets, the method achieves an average 6.2% AUC improvement over state-of-the-art baselines and demonstrates significantly enhanced robustness to noise and non-stationary dynamics.

Technology Category

• Machine Learning: Causal Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Reasoning under Uncertainty: Causality

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Utilizing the complex inter-variable causal relationships within multivariate time-series provides a promising avenue toward more robust and reliable multivariate time-series anomaly detection (MTSAD) but remains an underexplored area of research. This paper proposes Causality-Aware contrastive learning for RObust multivariate Time-Series (CAROTS), a novel MTSAD pipeline that incorporates the notion of causality into contrastive learning. CAROTS employs two data augmentors to obtain causality-preserving and -disturbing samples that serve as a wide range of normal variations and synthetic anomalies, respectively. With causality-preserving and -disturbing samples as positives and negatives, CAROTS performs contrastive learning to train an encoder whose latent space separates normal and abnormal samples based on causality. Moreover, CAROTS introduces a similarity-filtered one-class contrastive loss that encourages the contrastive learning process to gradually incorporate more semantically diverse samples with common causal relationships. Extensive experiments on five real-world and two synthetic datasets validate that the integration of causal relationships endows CAROTS with improved MTSAD capabilities. The code is available at https://github.com/kimanki/CAROTS.

Problem

• Incorporates causality into contrastive learning for anomaly detection
• Uses causality-preserving and -disturbing samples for robust training
• Improves anomaly detection by leveraging inter-variable causal relationships

Innovation

• Uses causality-aware contrastive learning
• Employs causality-preserving and -disturbing samples
• Introduces similarity-filtered one-class contrastive loss