When Model Meets New Normals: Test-time Adaptation for Unsupervised Time-series Anomaly Detection

TL;DR

This paper addresses the challenge of evolving normal patterns in time-series anomaly detection by introducing a test-time adaptation method that improves robustness to distribution shifts, validated through extensive real-world experiments.

Contribution

It proposes a novel test-time adaptation strategy using trend estimation and self-supervised learning to handle the new normal problem in unsupervised time-series anomaly detection.

Findings

Improved detection performance on real-world benchmarks.

Enhanced robustness to distribution shifts.

Consistent performance gains over baseline methods.

Abstract

Time-series anomaly detection deals with the problem of detecting anomalous timesteps by learning normality from the sequence of observations. However, the concept of normality evolves over time, leading to a "new normal problem", where the distribution of normality can be changed due to the distribution shifts between training and test data. This paper highlights the prevalence of the new normal problem in unsupervised time-series anomaly detection studies. To tackle this issue, we propose a simple yet effective test-time adaptation strategy based on trend estimation and a self-supervised approach to learning new normalities during inference. Extensive experiments on real-world benchmarks demonstrate that incorporating the proposed strategy into the anomaly detector consistently improves the model's performance compared to the baselines, leading to robustness to the distribution shifts.