An Encode-then-Decompose Approach to Unsupervised Time Series Anomaly Detection on Contaminated Training Data--Extended Version

TL;DR

This paper introduces an encode-then-decompose method for unsupervised time series anomaly detection that improves robustness against contaminated training data, outperforming existing approaches on multiple benchmarks.

Contribution

The paper proposes a novel encode-then-decompose framework and a mutual information-based anomaly metric, enhancing robustness and accuracy in contaminated training scenarios.

Findings

Achieves state-of-the-art performance on eight benchmarks.

Demonstrates robustness across different contamination ratios.

Outperforms traditional autoencoder-based methods.

Abstract

Time series anomaly detection is important in modern large-scale systems and is applied in a variety of domains to analyze and monitor the operation of diverse systems. Unsupervised approaches have received widespread interest, as they do not require anomaly labels during training, thus avoiding potentially high costs and having wider applications. Among these, autoencoders have received extensive attention. They use reconstruction errors from compressed representations to define anomaly scores. However, representations learned by autoencoders are sensitive to anomalies in training time series, causing reduced accuracy. We propose a novel encode-then-decompose paradigm, where we decompose the encoded representation into stable and auxiliary representations, thereby enhancing the robustness when training with contaminated time series. In addition, we propose a novel mutual information based metric to replace the reconstruction errors for identifying anomalies. Our proposal demonstrates competitive or state-of-the-art performance on eight commonly used multi- and univariate time series benchmarks and exhibits robustness to time series with different contamination ratios.