DDMT: Denoising Diffusion Mask Transformer Models for Multivariate Time Series Anomaly Detection

TL;DR

This paper introduces DDMT, a novel framework combining Denoising Diffusion Models and Transformers with an adaptive neighbor mask to improve multivariate time series anomaly detection, effectively handling noise and information leakage.

Contribution

The paper presents the first integration of Denoising Diffusion Models with Transformers for multivariate time series anomaly detection, introducing the ADNM mechanism to reduce information leakage.

Findings

Achieves state-of-the-art anomaly detection performance on five datasets.

Effectively mitigates noise and Weak Identity Mapping issues during reconstruction.

Demonstrates robustness and accuracy in complex multivariate time series data.

Abstract

Anomaly detection in multivariate time series has emerged as a crucial challenge in time series research, with significant research implications in various fields such as fraud detection, fault diagnosis, and system state estimation. Reconstruction-based models have shown promising potential in recent years for detecting anomalies in time series data. However, due to the rapid increase in data scale and dimensionality, the issues of noise and Weak Identity Mapping (WIM) during time series reconstruction have become increasingly pronounced. To address this, we introduce a novel Adaptive Dynamic Neighbor Mask (ADNM) mechanism and integrate it with the Transformer and Denoising Diffusion Model, creating a new framework for multivariate time series anomaly detection, named Denoising Diffusion Mask Transformer (DDMT). The ADNM module is introduced to mitigate information leakage between input and output features during data reconstruction, thereby alleviating the problem of WIM during reconstruction. The Denoising Diffusion Transformer (DDT) employs the Transformer as an internal neural network structure for Denoising Diffusion Model. It learns the stepwise generation process of time series data to model the probability distribution of the data, capturing normal data patterns and progressively restoring time series data by removing noise, resulting in a clear recovery of anomalies. To the best of our knowledge, this is the first model that combines Denoising Diffusion Model and the Transformer for multivariate time series anomaly detection. Experimental evaluations were conducted on five publicly available multivariate time series anomaly detection datasets. The results demonstrate that the model effectively identifies anomalies in time series data, achieving state-of-the-art performance in anomaly detection.