Back to Repair: A Minimal Denoising Network for Time Series Anomaly Detection

TL;DR

JuRe is a minimalistic denoising network for time series anomaly detection that achieves competitive results without complex architecture components, emphasizing the importance of the denoising objective.

Contribution

The paper introduces JuRe, a simple yet effective denoising network that challenges the need for architectural complexity in time series anomaly detection.

Findings

JuRe ranks second on TSB-AD benchmark with an AUC-PR of 0.404.

JuRe ranks second on UCR univariate archive with an AUC-PR of 0.198.

Training-time corruption is the key factor influencing detection performance.

Abstract

We introduce JuRe (Just Repair), a minimal denoising network for time series anomaly detection that exposes a central finding: architectural complexity is unnecessary when the training objective correctly implements the manifold-projection principle. JuRe consists of a single depthwise-separable convolutional residual block with hidden dimension 128, trained to repair corrupted time series windows and scored at inference by a fixed, parameter-free structural discrepancy function. Despite using no attention, no latent variable, and no adversarial component, JuRe ranks second on the TSB-AD multivariate benchmark (AUC-PR 0.404, 180 series, 17 datasets) and second on the UCR univariate archive by AUC-PR (0.198, 250 series), leading all neural baselines on AUC-PR and VUS-PR. Component ablation on TSB-AD identifies training-time corruption as the dominant factor ($\Delta$AUC-PR $= 0.047$ on removal), confirming that the denoising objective, not network capacity, drives detection quality. Pairwise Wilcoxon signed-rank tests establish statistical significance against 21 of 25 baselines on TSB-AD. Code is available at the URL https://github.com/iis-esslingen/JuRe.