Explainable Time Series Anomaly Detection using Masked Latent Generative Modeling

Abstract

We present a novel time series anomaly detection method that achieves excellent detection accuracy while offering a superior level of explainability. Our proposed method, TimeVQVAE-AD, leverages masked generative modeling adapted from the cutting-edge time series generation method known as TimeVQVAE. The prior model is trained on the discrete latent space of a time-frequency domain. Notably, the dimensional semantics of the time-frequency domain are preserved in the latent space, enabling us to compute anomaly scores across different frequency bands, which provides a better insight into the detected anomalies. Additionally, the generative nature of the prior model allows for sampling likely normal states for detected anomalies, enhancing the explainability of the detected anomalies through counterfactuals. Our experimental evaluation on the UCR Time Series Anomaly archive demonstrates that TimeVQVAE-AD significantly surpasses the existing methods in terms of detection accuracy and explainability. We provide our implementation on GitHub: https://github.com/ML4ITS/TimeVQVAE-AnomalyDetection.

Published in Pattern Recognition

References in corpus (10)

• Neural Discrete Representation Learning
• Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding
• Time-Series Anomaly Detection Service at Microsoft
• Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progress
• TFAD: A Decomposition Time Series Anomaly Detection Architecture with Time-Frequency Analysis
• Detecting Regions of Maximal Divergence for Spatio-Temporal Anomaly Detection
• Is it worth it? Comparing six deep and classical methods for unsupervised anomaly detection in time series
• Merlion: A Machine Learning Library for Time Series
• Contrastive Predictive Coding for Anomaly Detection
• Diverse Counterfactual Explanations for Anomaly Detection in Time Series

Cited by in corpus (2)

• Latent Diffusion Model for Conditional Reservoir Facies Generation
• DACAD: Domain Adaptation Contrastive Learning for Anomaly Detection in Multivariate Time Series