Multidimensional Time Series Anomaly Detection: A GRU-based Gaussian Mixture Variational Autoencoder Approach

Yifan Guo, Weixian Liao, Qianlong Wang, Lixing Yu, Tianxi Ji, Pan Li
Proceedings of The 10th Asian Conference on Machine Learning, PMLR 95:97-112, 2018.

Abstract

Unsupervised anomaly detection on multidimensional time series data is a very important problem due to its wide applications in many systems such as cyber-physical systems, the Internet of Things. Some existing works use traditional variational autoencoder (VAE) for anomaly detection. They generally assume a single-modal Gaussian distribution as prior in the data generative procedure. However, because of the intrinsic multimodality in time series data, previous works cannot effectively learn the complex data distribution, and hence cannot make accurate detections. To tackle this challenge, in this paper, we propose a GRU-based Gaussian Mixture VAE system for anomaly detection, called GGM-VAE. In particular, Gated Recurrent Unit (GRU) cells are employed to discover the correlations among time sequences. Then we use Gaussian Mixture priors in the latent space to characterize multimodal data. The proposed detector reports an anomaly when the reconstruction probability is below a certain threshold. We conduct extensive simulations on real world datasets and find that our proposed scheme outperforms the state-of-the-art anomaly detection schemes and achieves up to 5.7% and 7.2% improvements in accuracy and F1 score, respectively, compared with existing methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v95-guo18a, title = {Multidimensional Time Series Anomaly Detection: A GRU-based Gaussian Mixture Variational Autoencoder Approach}, author = {Guo, Yifan and Liao, Weixian and Wang, Qianlong and Yu, Lixing and Ji, Tianxi and Li, Pan}, booktitle = {Proceedings of The 10th Asian Conference on Machine Learning}, pages = {97--112}, year = {2018}, editor = {Zhu, Jun and Takeuchi, Ichiro}, volume = {95}, series = {Proceedings of Machine Learning Research}, month = {14--16 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v95/guo18a/guo18a.pdf}, url = {https://proceedings.mlr.press/v95/guo18a.html}, abstract = {Unsupervised anomaly detection on multidimensional time series data is a very important problem due to its wide applications in many systems such as cyber-physical systems, the Internet of Things. Some existing works use traditional variational autoencoder (VAE) for anomaly detection. They generally assume a single-modal Gaussian distribution as prior in the data generative procedure. However, because of the intrinsic multimodality in time series data, previous works cannot effectively learn the complex data distribution, and hence cannot make accurate detections. To tackle this challenge, in this paper, we propose a GRU-based Gaussian Mixture VAE system for anomaly detection, called GGM-VAE. In particular, Gated Recurrent Unit (GRU) cells are employed to discover the correlations among time sequences. Then we use Gaussian Mixture priors in the latent space to characterize multimodal data. The proposed detector reports an anomaly when the reconstruction probability is below a certain threshold. We conduct extensive simulations on real world datasets and find that our proposed scheme outperforms the state-of-the-art anomaly detection schemes and achieves up to 5.7% and 7.2% improvements in accuracy and F1 score, respectively, compared with existing methods.} }
Endnote
%0 Conference Paper %T Multidimensional Time Series Anomaly Detection: A GRU-based Gaussian Mixture Variational Autoencoder Approach %A Yifan Guo %A Weixian Liao %A Qianlong Wang %A Lixing Yu %A Tianxi Ji %A Pan Li %B Proceedings of The 10th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jun Zhu %E Ichiro Takeuchi %F pmlr-v95-guo18a %I PMLR %P 97--112 %U https://proceedings.mlr.press/v95/guo18a.html %V 95 %X Unsupervised anomaly detection on multidimensional time series data is a very important problem due to its wide applications in many systems such as cyber-physical systems, the Internet of Things. Some existing works use traditional variational autoencoder (VAE) for anomaly detection. They generally assume a single-modal Gaussian distribution as prior in the data generative procedure. However, because of the intrinsic multimodality in time series data, previous works cannot effectively learn the complex data distribution, and hence cannot make accurate detections. To tackle this challenge, in this paper, we propose a GRU-based Gaussian Mixture VAE system for anomaly detection, called GGM-VAE. In particular, Gated Recurrent Unit (GRU) cells are employed to discover the correlations among time sequences. Then we use Gaussian Mixture priors in the latent space to characterize multimodal data. The proposed detector reports an anomaly when the reconstruction probability is below a certain threshold. We conduct extensive simulations on real world datasets and find that our proposed scheme outperforms the state-of-the-art anomaly detection schemes and achieves up to 5.7% and 7.2% improvements in accuracy and F1 score, respectively, compared with existing methods.
APA
Guo, Y., Liao, W., Wang, Q., Yu, L., Ji, T. & Li, P.. (2018). Multidimensional Time Series Anomaly Detection: A GRU-based Gaussian Mixture Variational Autoencoder Approach. Proceedings of The 10th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 95:97-112 Available from https://proceedings.mlr.press/v95/guo18a.html.

Related Material