Anomaly Detection by Recombining Gated Unsupervised Experts

被引:0
作者
Schulze, Jan-Philipp [1 ,2 ]
Sperl, Philip [1 ,2 ]
Boettinger, Konstantin [2 ]
机构
[1] Tech Univ Munich, Munich, Germany
[2] Fraunhofer Inst Appl & Integrated Secur, Garching, Germany
来源
2022 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN) | 2022年
关键词
anomaly detection; deep learning; unsupervised learning; data mining; mixture-of-experts; IT security; MIXTURE;
D O I
10.1109/IJCNN55064.2022.9892807
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Anomaly detection has been considered under several extents of prior knowledge. Unsupervised methods do not require any labelled data, whereas semi-supervised methods leverage some known anomalies. Inspired by mixture-of-experts models and the analysis of the hidden activations of neural networks, we introduce a novel data-driven anomaly detection method called ARGUE. Our method is not only applicable to unsupervised and semi-supervised environments, but also profits from prior knowledge of self-supervised settings. We designed ARGUE as a combination of dedicated expert networks, which specialise on parts of the input data. For its final decision, ARGUE fuses the distributed knowledge across the expert systems using a gated mixture-of-experts architecture. Our evaluation motivates that prior knowledge about the normal data distribution may be as valuable as known anomalies.
引用
收藏
页数:8
相关论文
共 48 条
[1]   GANomaly: Semi-supervised Anomaly Detection via Adversarial Training [J].
Akcay, Samet ;
Atapour-Abarghouei, Amir ;
Breckon, Toby P. .
COMPUTER VISION - ACCV 2018, PT III, 2019, 11363 :622-637
[2]  
[Anonymous], 2023, 2009 IEEE S COMP INT
[3]  
Bhatia S, 2020, AAAI CONF ARTIF INTE, V34, P3243
[4]   Comparative accuracies of artificial neural networks and discriminant analysis in predicting forest cover types from cartographic variables [J].
Blackard, JA ;
Dean, DJ .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 1999, 24 (03) :131-151
[5]  
Borghesi A, 2019, AAAI CONF ARTIF INTE, P9428
[6]   Support vector machines experts for time series forecasting [J].
Cao, LJ .
NEUROCOMPUTING, 2003, 51 :321-339
[7]   CAE: Contextual auto-encoder for multivariate time-series anomaly detection in air transportation [J].
Chevrot, Antoine ;
Vernotte, Alexandre ;
Legeard, Bruno .
COMPUTERS & SECURITY, 2022, 116
[8]  
Cohen G, 2017, IEEE IJCNN, P2921, DOI 10.1109/IJCNN.2017.7966217
[9]  
Dua D, 2017, UCI machine learning repository
[10]  
Guo J, 2018, 2018 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING (EMNLP 2018), P4694