Outlier Detection for Time Series with Recurrent Autoencoder Ensembles

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

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

Abstract

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.
Original languageEnglish
Title of host publicationProceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019
EditorsSarit Kraus
Number of pages8
Publisherijcai.org
Publication date2019
Pages2725-2732
DOIs
Publication statusPublished - 2019
Eventthe 28th International Joint Conference on Artificial Intelligence - Macau, China
Duration: 10 Aug 2019 → …
https://www.ijcai19.org/

Conference

Conferencethe 28th International Joint Conference on Artificial Intelligence
CountryChina
CityMacau
Period10/08/2019 → …
Internet address

Fingerprint

Time series
Recurrent neural networks
Neural networks
Experiments

Cite this

Kieu, T., Yang, B., Guo, C., & Jensen, C. S. (2019). Outlier Detection for Time Series with Recurrent Autoencoder Ensembles. In S. Kraus (Ed.), Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019 (pp. 2725-2732). ijcai.org. https://doi.org/10.24963/ijcai.2019/378
Kieu, Tung ; Yang, Bin ; Guo, Chenjuan ; Jensen, Christian S. / Outlier Detection for Time Series with Recurrent Autoencoder Ensembles. Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019. editor / Sarit Kraus. ijcai.org, 2019. pp. 2725-2732
@inproceedings{f1d172c18bda4d83addc597ac8637cbb,
title = "Outlier Detection for Time Series with Recurrent Autoencoder Ensembles",
abstract = "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.",
author = "Tung Kieu and Bin Yang and Chenjuan Guo and Jensen, {Christian S.}",
year = "2019",
doi = "10.24963/ijcai.2019/378",
language = "English",
pages = "2725--2732",
editor = "Sarit Kraus",
booktitle = "Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019",
publisher = "ijcai.org",

}

Kieu, T, Yang, B, Guo, C & Jensen, CS 2019, Outlier Detection for Time Series with Recurrent Autoencoder Ensembles. in S Kraus (ed.), Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019. ijcai.org, pp. 2725-2732, the 28th International Joint Conference on Artificial Intelligence, Macau, China, 10/08/2019. https://doi.org/10.24963/ijcai.2019/378

Outlier Detection for Time Series with Recurrent Autoencoder Ensembles. / Kieu, Tung; Yang, Bin; Guo, Chenjuan; Jensen, Christian S.

Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019. ed. / Sarit Kraus. ijcai.org, 2019. p. 2725-2732.

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

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.

U2 - 10.24963/ijcai.2019/378

DO - 10.24963/ijcai.2019/378

M3 - Article in proceeding

SP - 2725

EP - 2732

BT - Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019

A2 - Kraus, Sarit

PB - ijcai.org

ER -

Kieu T, Yang B, Guo C, Jensen CS. Outlier Detection for Time Series with Recurrent Autoencoder Ensembles. In Kraus S, editor, Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019. ijcai.org. 2019. p. 2725-2732 https://doi.org/10.24963/ijcai.2019/378

Access to Document