Outlier Detection for Time Series with Recurrent Autoencoder Ensembles

Outlier Detection for Time Series with Recurrent Autoencoder Ensembles

Tung Kieu, Bin Yang, Chenjuan Guo, Christian S. Jensen

Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence
Main track. Pages 2725-2732. https://doi.org/10.24963/ijcai.2019/378

We propose two solutions to outlier detection in time series based on recurrent autoencoder ensembles. The solutions exploit autoencoders built using sparsely-connected recurrent neural networks (S-RNNs). Such networks make it possible to generate multiple autoencoders with different neural network connection structures. The two solutions are ensemble frameworks, specifically an independent framework and a shared framework, both of which combine multiple S-RNN based autoencoders to enable outlier detection.  This ensemble-based approach aims to reduce the effects of some autoencoders being overfitted to outliers, this way improving overall detection quality. Experiments with two large real-world time series data sets, including univariate and multivariate time series, offer insight into the design properties of the proposed frameworks and demonstrate that the resulting solutions are capable of outperforming both baselines and the state-of-the-art methods.
Keywords:
Machine Learning: Data Mining
Machine Learning: Ensemble Methods
Machine Learning: Time-series;Data Streams
Machine Learning: Unsupervised Learning
Machine Learning Applications: Applications of Unsupervised Learning