Time Series Anomaly Detection With Adversarial Reconstruction Networks
- 4 January 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 35 (4) , 4293-4306
Abstract
Time series data naturally exist in many domains including medical data analysis, infrastructure sensor monitoring, and motion tracking. However, a very small portion of anomalous time series can be observed, comparing to the whole data. Most existing approaches are based on the supervised classification model requiring representative labels for anomaly class(es), which is challenging in real-world problems. So can we learn how to detect anomalous time ticks in an effective yet efficient way, given mostly normal time series data? Therefore, we propose an unsupervised reconstruction model named BeatGAN which learns to detect anomalies based on normal data, or data which majority of samples are normal. BeatGAN provides a framework to adversarially learn to reconstruct, which can cooperate with both 1-d CNN and RNN. Rarely observed anomalies can result in larger reconstruction errors, which are then detected based on extreme value theory. Moreover, data augmentation with dynamic time warping regularizes reconstruction and provides robustness. In the experiments, effectiveness and sensitivity are studied in both synthetic data and various real-world time series. BeatGAN achieves better accuracy and fast inference.Keywords
Funding Information
- National Natural Science Foundation of China (61772498, 61872206, 91746301, U1911401)
- Strategic Priority Research Program of the Chinese Academy of Sciences (XDA19020400)
This publication has 36 references indexed in Scilit:
- SWaT: a water treatment testbed for research and training on ICS security2016
- Real-Time Patient-Specific ECG Classification by 1-D Convolutional Neural NetworksIEEE Transactions on Biomedical Engineering, 2015
- Spotting Suspicious Link Behavior with fBox: An Adversarial Perspective2014
- AutoPlait2014
- Quantile estimation for the generalized pareto distribution with application to finance2012
- Anomaly detectionACM Computing Surveys, 2009
- The impact of the MIT-BIH Arrhythmia DatabaseIEEE Engineering in Medicine and Biology Magazine, 2001
- Speech discrimination by dynamic programmingCybernetics and Systems Analysis, 1972
- Statistical Inference for Probabilistic Functions of Finite State Markov ChainsThe Annals of Mathematical Statistics, 1966
- Sample Criteria for Testing Outlying ObservationsThe Annals of Mathematical Statistics, 1950