MLP-Mixer based Masked Autoencoders Are Effective, Explainable and Robust for Time Series Anomaly Detection

Download PDF

Abstract

Time series anomaly detection remains one of the most active re-Time series anomaly detection remains one of the most active research areas in data mining due to its wide range of real-world ap-search areas in data mining due to its wide range of real-world applications. In recent years, numerous deep learning-based methods have been proposed for this task. However, deep learning-based methods fail to detect subsequence anomalies with long durations, lack explainability, and are vulnerable to training set contamina-lack explainability, and are vulnerable to training set contamination. This paper addresses these issues by proposing a novel deep learning framework for effective, explainable, and robust time se-learning framework for effective, explainable, and robust time series anomaly detection. Our framework, MMA , incorporates the M LP-Mixer backbone with a M asked A utoencoder-based anomaly detection approach to allow for a significantly larger input win-detection approach to allow for a significantly larger input window size (10 to 20 times larger than the input window sizes of cur-dow size (10 to 20 times larger than the input window sizes of current models). This larger input window enables our model to detect challenging subsequence anomalies. Meanwhile, a contrast learn-challenging subsequence anomalies. Meanwhile, a contrast learning module is proposed to aid in detecting subtle anomalies that fail to be identified by residual errors. Furthermore, a dynamic anom-to be identified by residual errors. Furthermore, a dynamic anomaly filtering method is introduced to mitigate the impact of sub-aly filtering method is introduced to mitigate the impact of subsequence anomalies on the reconstruction of surrounding normal regions to reduce false alarms. Extensive experiments on univari-regions to reduce false alarms. Extensive experiments on univariate and multivariate time series datasets demonstrate that our pro-ate and multivariate time series datasets demonstrate that our proposed framework significantly outperforms state-of-the-art meth-posed framework significantly outperforms state-of-the-art methods across rigorous evaluation metrics. Additionally, MMA has a strong ability to reconstruct potential normal patterns in anoma-strong ability to reconstruct potential normal patterns in anomalous regions, providing high levels of explainability. Moreover, MMA demonstrates high robustness to various types of training set pollution.