Mamba Adaptive Anomaly Transformer with association discrepancy for time series

TL;DR

This paper introduces MAAT, an advanced anomaly detection model for time series that improves association modeling and reconstruction, leading to better detection accuracy in complex, noisy environments.

Contribution

The paper presents MAAT, a novel architecture combining Sparse Attention and Mamba-Selective State Space Model for enhanced anomaly detection in time series.

Findings

MAAT outperforms existing methods in anomaly detection accuracy.

MAAT demonstrates strong generalization across diverse time series datasets.

The approach effectively handles noisy and non-stationary data environments.

Abstract

Anomaly detection in time series is essential for industrial monitoring and environmental sensing, yet distinguishing anomalies from complex patterns remains challenging. Existing methods like the Anomaly Transformer and DCdetector have progressed, but they face limitations such as sensitivity to short-term contexts and inefficiency in noisy, non-stationary environments. To overcome these issues, we introduce MAAT, an improved architecture that enhances association discrepancy modeling and reconstruction quality. MAAT features Sparse Attention, efficiently capturing long-range dependencies by focusing on relevant time steps, thereby reducing computational redundancy. Additionally, a Mamba-Selective State Space Model is incorporated into the reconstruction module, utilizing a skip connection and Gated Attention to improve anomaly localization and detection performance. Extensive experiments show that MAAT significantly outperforms previous methods, achieving better anomaly distinguishability and generalization across various time series applications, setting a new standard for unsupervised time series anomaly detection in real-world scenarios.