Calibrated One-class Classification for Unsupervised Time Series Anomaly Detection

作     者：Xu, Hongzuo Wang, Yijie Jian, Songlei Liao, Qing Wang, Yongjun Pang, Guansong 

作者机构： Beijing100091 China The National Key Laboratory of Parallel and Distributed Computing College of Computer National University of Defense Technology Hunan410073 China The College of Computer National University of Defense Technology Hunan410073 China  Guangdong518055 China The School of Computing and Information Systems Singapore Management University Singapore178902 Singapore 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2022年

核心收录：

主　　题：Contamination 

摘      要：Time series anomaly detection is instrumental in maintaining system availability in various domains. Current work in this research line mainly focuses on learning data normality deeply and comprehensively by devising advanced neural network structures and new reconstruction/prediction learning objectives. However, their one-class learning process can be misled by latent anomalies in training data (i.e., anomaly contamination) under the unsupervised paradigm. Their learning process also lacks knowledge about the anomalies. Consequently, they often learn a biased, inaccurate normality boundary. To tackle these problems, this paper proposes calibrated one-class classification for anomaly detection, realizing contamination-tolerant, anomaly-informed learning of data normality via uncertainty modeling-based calibration and native anomaly-based calibration. Specifically, our approach adaptively penalizes uncertain predictions to restrain irregular samples in anomaly contamination during optimization, while simultaneously encouraging confident predictions on regular samples to ensure effective normality learning. This largely alleviates the negative impact of anomaly contamination. Our approach also creates native anomaly examples via perturbation to simulate time series abnormal behaviors. Through discriminating these dummy anomalies, our one-class learning is further calibrated to form a more precise normality boundary. Extensive experiments on ten real-world datasets show that our model achieves substantial improvement over sixteen state-of-the-art contenders. © 2022, CC BY.