CGSTA: Cross-Scale Graph Contrast with Stability-Aware Alignment for Multivariate Time-Series Anomaly Detection

TL;DR

CGSTA introduces a multi-scale, stability-aware graph contrastive framework for multivariate time-series anomaly detection, effectively handling evolving dependencies and noise to improve detection accuracy across multiple benchmarks.

Contribution

It proposes a novel multi-scale graph contrastive learning approach with stability-aware alignment to enhance anomaly detection in noisy, evolving multivariate time-series data.

Findings

Achieves state-of-the-art results on PSM and WADI datasets.

Performs comparably to baselines on SWaT and SMAP.

Effectively suppresses noise and captures dependencies across scales.

Abstract

Multivariate time-series anomaly detection is essential for reliable industrial control, telemetry, and service monitoring. However, the evolving inter-variable dependencies and inevitable noise render it challenging. Existing methods often use single-scale graphs or instance-level contrast. Moreover, learned dynamic graphs can overfit noise without a stable anchor, causing false alarms or misses. To address these challenges, we propose the CGSTA framework with two key innovations. First, Dynamic Layered Graph Construction (DLGC) forms local, regional, and global views of variable relations for each sliding window; rather than contrasting whole windows, Contrastive Discrimination across Scales (CDS) contrasts graph representations within each view and aligns the same window across views to make learning structure-aware. Second, Stability-Aware Alignment (SAA) maintains a per-scale stable reference learned from normal data and guides the current window's fast-changing graphs toward it to suppress noise. We fuse the multi-scale and temporal features and use a conditional density estimator to produce per-time-step anomaly scores. Across four benchmarks, CGSTA delivers optimal performance on PSM and WADI, and is comparable to the baseline methods on SWaT and SMAP.