Memory Mechanism for Unsupervised Anomaly Detection

Jiahao Li, Yiqiang Chen, Yunbing Xing
Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence, PMLR 216:1219-1229, 2023.

Abstract

Unsupervised anomaly detection is a binary classification that detects anomalies in unseen samples given only unlabeled normal data. Reconstruction-based approaches are widely used, which perform reconstruction error minimization on training data to learn normal patterns and quantify the degree of anomalies by reconstruction errors on testing data. However, this approach tends to miss anomalies when the normal data has multi-pattern. Because the model generalizes unrestrictedly beyond normal patterns even to include anomaly patterns. In this paper, we proposed a memory mechanism that memorizes typical normal patterns through a capacity-controlled external differentiable matrix so that the generalization of the model to anomalies is limited by the retrieval of the matrix. We achieved state-of-the-art performance on several public benchmarks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v216-li23a, title = {Memory Mechanism for Unsupervised Anomaly Detection}, author = {Li, Jiahao and Chen, Yiqiang and Xing, Yunbing}, booktitle = {Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence}, pages = {1219--1229}, year = {2023}, editor = {Evans, Robin J. and Shpitser, Ilya}, volume = {216}, series = {Proceedings of Machine Learning Research}, month = {31 Jul--04 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v216/li23a/li23a.pdf}, url = {https://proceedings.mlr.press/v216/li23a.html}, abstract = {Unsupervised anomaly detection is a binary classification that detects anomalies in unseen samples given only unlabeled normal data. Reconstruction-based approaches are widely used, which perform reconstruction error minimization on training data to learn normal patterns and quantify the degree of anomalies by reconstruction errors on testing data. However, this approach tends to miss anomalies when the normal data has multi-pattern. Because the model generalizes unrestrictedly beyond normal patterns even to include anomaly patterns. In this paper, we proposed a memory mechanism that memorizes typical normal patterns through a capacity-controlled external differentiable matrix so that the generalization of the model to anomalies is limited by the retrieval of the matrix. We achieved state-of-the-art performance on several public benchmarks.} }
Endnote
%0 Conference Paper %T Memory Mechanism for Unsupervised Anomaly Detection %A Jiahao Li %A Yiqiang Chen %A Yunbing Xing %B Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2023 %E Robin J. Evans %E Ilya Shpitser %F pmlr-v216-li23a %I PMLR %P 1219--1229 %U https://proceedings.mlr.press/v216/li23a.html %V 216 %X Unsupervised anomaly detection is a binary classification that detects anomalies in unseen samples given only unlabeled normal data. Reconstruction-based approaches are widely used, which perform reconstruction error minimization on training data to learn normal patterns and quantify the degree of anomalies by reconstruction errors on testing data. However, this approach tends to miss anomalies when the normal data has multi-pattern. Because the model generalizes unrestrictedly beyond normal patterns even to include anomaly patterns. In this paper, we proposed a memory mechanism that memorizes typical normal patterns through a capacity-controlled external differentiable matrix so that the generalization of the model to anomalies is limited by the retrieval of the matrix. We achieved state-of-the-art performance on several public benchmarks.
APA
Li, J., Chen, Y. & Xing, Y.. (2023). Memory Mechanism for Unsupervised Anomaly Detection. Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 216:1219-1229 Available from https://proceedings.mlr.press/v216/li23a.html.

Related Material