PAI: Preserving Amplitude Information in Representation-Based Time-Series Anomaly Detection

2026-06-07

AI Summary

This study addresses a critical limitation in existing representation-based time series anomaly detection methods, which often neglect amplitude information, leading to insufficient sensitivity to amplitude-related anomalies. To overcome this, the authors propose the PAI framework—the first approach to explicitly incorporate amplitude preservation into the representation learning pipeline. PAI introduces a diagnostic module to evaluate whether embeddings retain amplitude characteristics and enhances raw anomaly scores by fusing point-wise median/MAD deviation with local mean shift metrics. The framework seamlessly integrates with mainstream methods such as TS2Vec and PaAno, consistently improving all baseline performance across the TSB-AD-U-Eva and TAB UV benchmarks. Notably, it achieves average VUS-PR gains of 98.4% and 36.8%, respectively, with PaAno+PAI surpassing the current state-of-the-art by 15%.

Abstract

Representation-based time-series anomaly detection algorithms significantly outperform other methods on diverse anomaly detection tasks. However, we notice that they suffer from a major limitation in our evaluation - their learned embeddings are often amplitude-agnostic. Losing amplitude information can degrade performance on amplitude related anomalies, and this failure is prevalent across all existing representation-based methods. To address aforementioned issues, we propose a new anomaly scoring scheme named PAI. PAI consists of two complementary modules, a diagnostic module and a final score augmentation function. The diagnostic module compares cosine and Euclidean scoring on the same representation bank to test whether amplitude information is already captured in the learned representation. Then in final score augmentation function, PAI computes a point-wise median and MAD deviation score and a local mean-shift score-which are fused with the representation score to produce the final anomaly score. On the TSB-AD-U-Eva and TAB UV datasets, PAI improves all four evaluated representation-based methods across every reported metric, achieving average VUS-PR gains of 98.4% and 36.8%, respectively. Among all evaluated combinations, PaAno + PAI achieves the best performance, outperforming the state-of-the-art method by 15%. Further evaluation on bootstrap confidence intervals, anomaly-type breakdowns, and a TS2Vec input-normalization ablation further support the proposed scheme. These results suggest that explicitly retaining amplitude information is important for representation-based time-series anomaly detection, which has been underemphasized in existing scoring schemes. Code is available at: https://github.com/pantheon5100/PAI

Problem

• time-series anomaly detection
• representation-based methods
• amplitude information
• anomaly scoring

Innovation

• amplitude-aware anomaly detection
• representation-based time series
• anomaly scoring fusion
• diagnostic module
• score augmentation