VISTA: Unsupervised 2D Temporal Dependency Representations for Time Series Anomaly Detection
This work solves the problem of detecting anomalies in noisy, real-world time series data for industrial applications, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of time series anomaly detection by introducing VISTA, an unsupervised algorithm that addresses noise sensitivity and inadequate temporal dependency modeling in existing methods, achieving state-of-the-art performance on five multivariate benchmarks.
Time Series Anomaly Detection (TSAD) is essential for uncovering rare and potentially harmful events in unlabeled time series data. Existing methods are highly dependent on clean, high-quality inputs, making them susceptible to noise and real-world imperfections. Additionally, intricate temporal relationships in time series data are often inadequately captured in traditional 1D representations, leading to suboptimal modeling of dependencies. We introduce VISTA, a training-free, unsupervised TSAD algorithm designed to overcome these challenges. VISTA features three core modules: 1) Time Series Decomposition using Seasonal and Trend Decomposition via Loess (STL) to decompose noisy time series into trend, seasonal, and residual components; 2) Temporal Self-Attention, which transforms 1D time series into 2D temporal correlation matrices for richer dependency modeling and anomaly detection; and 3) Multivariate Temporal Aggregation, which uses a pretrained feature extractor to integrate cross-variable information into a unified, memory-efficient representation. VISTA's training-free approach enables rapid deployment and easy hyperparameter tuning, making it suitable for industrial applications. It achieves state-of-the-art performance on five multivariate TSAD benchmarks.