Bayesian autoencoders with uncertainty quantification: Towards trustworthy anomaly detection

Despite numerous studies of deep autoencoders (AEs) for unsupervised anomaly detection, AEs still lack a way to express uncertainty in their predictions, crucial for ensuring safe and trustworthy machine learning systems in high-stake applications. Therefore, in this work, the formulation of Bayesia...

Full description

Saved in:
Bibliographic Details
Published in:Expert systems with applications Vol. 209; p. 118196
Main Authors: Yong, Bang Xiang, Brintrup, Alexandra
Format: Journal Article
Language:English
Published: Elsevier Ltd 15.12.2022
Subjects:
Anomaly detection
Bayesian autoencoders
Trustworthy machine learning
Uncertainty
Uncertainty
Bayesian autoencoders
Anomaly detection
Trustworthy machine learning
ISSN:0957-4174, 1873-6793
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Despite numerous studies of deep autoencoders (AEs) for unsupervised anomaly detection, AEs still lack a way to express uncertainty in their predictions, crucial for ensuring safe and trustworthy machine learning systems in high-stake applications. Therefore, in this work, the formulation of Bayesian autoencoders (BAEs) is adopted to quantify the total anomaly uncertainty, comprising epistemic and aleatoric uncertainties. To evaluate the quality of uncertainty, we consider the task of classifying anomalies with the additional option of rejecting predictions of high uncertainty. In addition, we use the accuracy-rejection curve and propose the weighted average accuracy as a performance metric. Our experiments demonstrate the effectiveness of the BAE and total anomaly uncertainty on a set of benchmark datasets and two real datasets for manufacturing: one for condition monitoring, the other for quality inspection. •Formulation of Bayesian autoencoders is extended to quantify anomaly uncertainty.•The total anomaly uncertainty comprises epistemic and aleatoric uncertainties.•Rejection of predictions with high uncertainty improves performance.•Validation of proposed methods on multiple benchmarks and real use cases.
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2022.118196