TY - GEN
T1 - Outlier Detection for Time Series with Recurrent Autoencoder Ensembles
AU - Kieu, Tung
AU - Yang, Bin
AU - Guo, Chenjuan
AU - Jensen, Christian S.
PY - 2019
Y1 - 2019
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=85069505455&partnerID=8YFLogxK
U2 - 10.24963/ijcai.2019/378
DO - 10.24963/ijcai.2019/378
M3 - Article in proceeding
SP - 2725
EP - 2732
BT - Proceedings of the 28th International Joint Conference on Artificial Intelligence, IJCAI 2019
A2 - Kraus, Sarit
PB - ijcai.org
T2 - the 28th International Joint Conference on Artificial Intelligence
Y2 - 10 August 2019
ER -