FreCT: Frequency-Augmented Convolutional Transformer for Robust Time Series Anomaly Detection

Wenxin Zhang, Ding Xu, Guangzhen Yao, Xiaojian Lin, Renxiang Guan, Chengze Du, Renda Han, Xi Xuan, Cuicui Luo

Abstract: Time series anomaly detection is critical for system monitoring and risk identification across various domains. However, detecting anomalies remains a challenge for most reconstruction-based approaches. On the one hand, reconstruction-based techniques are susceptible to computational deviation stemming from anomalies, which can lead to impure representations of normal sequence patterns. On the other hand, they often focus on the time-domain dependencies of time series while ignoring the alignment of frequency information beyond the time domain. To address these challenges, we propose a novel Frequency-augmented Convolutional Transformer (FreCT). FreCT utilizes patch operations to generate contrastive views and employs an improved Transformer architecture with a convolution module to capture long-term dependencies while preserving local topology information. The introduced frequency analysis could enhance the model’s ability to capture crucial characteristics beyond the time domain. To improve the training quality, FreCT deploys stop-gradient Kullback-Leibler (KL) divergence and absolute error to optimize consistency information. Extensive experiments on four public datasets demonstrate that FreCT outperforms existing methods in identifying anomalies. The code is available at https://github.com/shaieesss/FreCT.

External IDs:dblp:conf/icic/ZhangXYLGDHXL25