Deep Contrastive One-Class Time Series Anomaly Detection

被引:0
|
作者
Wang, Rui [1 ]
Liu, Chongwei [1 ]
Mou, Xudong [1 ]
Gao, Kai [2 ]
Guo, Xiaohui [3 ]
Liu, Pin [4 ]
Wo, Tianyu [5 ]
Liu, Xudong [1 ,3 ]
机构
[1] Beihang Univ, Sch Comp Sci & Engn, Beijing, Peoples R China
[2] Univ New South Wales, Sydney, NSW, Australia
[3] Beihang Univ, Hangzhou Innovat Inst, Hangzhou, Peoples R China
[4] China Univ Geosci Beijing, Sch Informat Engn, Beijing, Peoples R China
[5] Beihang Univ, Coll Software, Beijing, Peoples R China
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The accumulation of time-series data and the absence of labels make time-series Anomaly Detection (AD) a self-supervised deep learning task. Single-normality-assumption-based methods, which reveal only a certain aspect of the whole normality, are incapable of tasks involved with a large number of anomalies. Specifically, Contrastive Learning (CL) methods distance negative pairs, many of which consist of both normal samples, thus reducing the AD performance. Existing multi-normality-assumption-based methods are usually two-staged, firstly pre-training through certain tasks whose target may differ from AD, limiting their performance. To overcome the shortcomings, a deep Contrastive One-Class Anomaly detection method of time series (COCA) is proposed by authors, following the normality assumptions of CL and one-class classification. It treats the original and reconstructed representations as the positive pair of negative-sample-free CL, namely "sequence contrast". Next, invariance terms and variance terms compose a contrastive one-class loss function in which the loss of the assumptions is optimized by invariance terms simultaneously and the "hypersphere collapse" is prevented by variance terms. In addition, extensive experiments on two real-world time-series datasets show the superior performance of the proposed method achieves state-of-the-art.
引用
收藏
页码:694 / 702
页数:9
相关论文
共 50 条
  • [1] Time Series Anomaly Detection Using Contrastive Learning based One-Class Classification
    Lee, Yeseul
    Byun, Yunseon
    Baek, Jun-Geol
    2023 INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE IN INFORMATION AND COMMUNICATION, ICAIIC, 2023, : 330 - 335
  • [2] Calibrated One-class Classification for Unsupervised Time Series Anomaly Detection
    Xu H.
    Wang Y.
    Jian S.
    Liao Q.
    Wang Y.
    Pang G.
    IEEE Transactions on Knowledge and Data Engineering, 2024, 36 (11) : 1 - 14
  • [3] A Deep One-class Model for Network Anomaly Detection
    Dai, Songlin
    Yan, Jubin
    Wang, Xiaoming
    Zhang, Lin
    2019 INTERNATIONAL CONFERENCE ON ADVANCED ELECTRONIC MATERIALS, COMPUTERS AND MATERIALS ENGINEERING (AEMCME 2019), 2019, 563
  • [4] Anomaly Detection using Clustered Deep One-Class Classification
    Kim, Younghwan
    Kim, Huy Kang
    2020 15TH ASIA JOINT CONFERENCE ON INFORMATION SECURITY (ASIAJCIS 2020), 2020, : 151 - 157
  • [5] Active anomaly detection based on deep one-class classification
    Kim, Minkyung
    Kim, Junsik
    Yu, Jongmin
    Choi, Jun Kyun
    PATTERN RECOGNITION LETTERS, 2023, 167 : 18 - 24
  • [6] Refining one-class representation: A unified transformer for unsupervised time-series anomaly detection
    Zhong, Guoxiang
    Liu, Fagui
    Jiang, Jun
    Wang, Bin
    Chen, C. L. Philip
    INFORMATION SCIENCES, 2024, 656
  • [7] Improving one-class SVM for anomaly detection
    Li, KL
    Huang, HK
    Tian, SF
    Xu, W
    2003 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-5, PROCEEDINGS, 2003, : 3077 - 3081
  • [8] A NEW ONE-CLASS SVM FOR ANOMALY DETECTION
    Chen, Yuting
    Qian, Jing
    Saligrama, Ventatesh
    2013 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2013, : 3567 - 3571
  • [9] Cluster-based Deep One-Class Classification Model for Anomaly Detection
    Kim, Younghwan
    Kim, Huy Kang
    JOURNAL OF INTERNET TECHNOLOGY, 2021, 22 (04): : 903 - 911
  • [10] Deep Domain-Adversarial Anomaly Detection With One-Class Transfer Learning
    Wentao Mao
    Gangsheng Wang
    Linlin Kou
    Xihui Liang
    IEEE/CAA Journal of Automatica Sinica, 2023, 10 (02) : 524 - 546