Disentangled Anomaly Detection For Multivariate Time Series

Xin Jie, Xixi Zhou, Chanfei Su, Zijun Zhou, Yuqing Yuan, Jiajun Bu, Haishuai Wang

Abstract

Anomaly detection in time series that aims to identify unusual patterns has attracted a lot of attention recently. However, the representation of abnormal and normal data is difffcult to be distinguished because they are usually entangled. Recently, disentanglement theory based on variational auto-encoder (VAE) has shown great potential in machine learning and achieved great success in computer vision and natural language processing. In this paper, we propose a novel disentangled anomaly detection approach that adopts VAE-based disentanglement networks for anomaly detection in multivariate time series. The proposed method learns highquality disentangled latent factors in a continuous representation space to facilitate the identiffcation of anomalies from normal data. Extensive experiments demonstrate that our proposed lightweight model DA-VAE achieves state-of-the-art performance.
BibTeX
@inproceedings{jie2024disentangled,
  author = {Xin Jie and Xixi Zhou and Chanfei Su and Zijun Zhou and Yuqing Yuan and Jiajun Bu and Haishuai Wang},
  title = {{Disentangled Anomaly Detection For Multivariate Time Series}},
  booktitle = {The Web Conference},
  year = {2024},
  doi = {10.1145/3589335.3651492},
}

Loading related content...

Figure