ARTA: Adversarial-Robust Multivariate Time--Series Anomaly Detection via Sparsity-Constrained Perturbations

TL;DR

This paper introduces ARTA, a framework that enhances multivariate time-series anomaly detection robustness against adversarial perturbations through joint training of a detector and a mask generator.

Contribution

The paper presents a novel adversarial training approach with a sparsity-constrained generator to improve robustness and interpretability of time-series anomaly detectors.

Findings

ARTA improves detection accuracy across multiple datasets.

The method results in more stable performance under increasing noise.

Generated masks explain detector sensitivity to corruptions.

Abstract

Time-series anomaly detection (TSAD) is a critical component in monitoring complex systems, yet modern deep learning-based detectors are often highly sensitive to localized input corruptions and structured noise. We propose ARTA (Adversarially Robust multivariate Time-series Anomaly detection via sparsity-constrained perturbations), a joint training framework that improves detector robustness through a principled min-max optimization objective. ARTA comprises an anomaly detector and a sparsity-constrained mask generator that are trained simultaneously. The generator identifies minimal, task-relevant temporal perturbations that maximally increase the detector's anomaly score, while the detector is optimized to remain stable under these structured perturbations. The resulting masks characterize the detector's sensitivity to adversarial temporal corruptions and can serve as explanatory signals for the detector's decisions. This adversarial training strategy exposes brittle decision pathways and encourages the detector to rely on distributed and stable temporal patterns rather than spurious localized artifacts. We conduct extensive experiments on the TSB-AD benchmark, demonstrating that ARTA consistently improves anomaly detection performance across diverse datasets and exhibits significantly more graceful degradation under increasing noise levels compared to state-of-the-art baselines.