Latent-Insensitive Autoencoders for Anomaly Detection

Mathematics

Artem A. Lenskiy¹  Muhammad S. Battikh² 

[1] School of Computing, The Australian National University, Canberra 2601, Australia;Systems and Computer Engineering Department, Al-Azhar University, Cairo 11651, Egypt;

关键词: anomaly detection;    autoencoders;    one-class classification;    principal components analysis;    self-taught learning;    negative examples;   

DOI  :  10.3390/math10010112

来源: DOAJ

【 摘 要 】

Reconstruction-based approaches to anomaly detection tend to fall short when applied to complex datasets with target classes that possess high inter-class variance. Similar to the idea of self-taught learning used in transfer learning, many domains are rich with similar unlabeled datasets that could be leveraged as a proxy for out-of-distribution samples. In this paper we introduce the latent-insensitive autoencoder (LIS-AE) where unlabeled data from a similar domain are utilized as negative examples to shape the latent layer (bottleneck) of a regular autoencoder such that it is only capable of reconstructing one task. We provide theoretical justification for the proposed training process and loss functions along with an extensive ablation study highlighting important aspects of our model. We test our model in multiple anomaly detection settings presenting quantitative and qualitative analysis showcasing the significant performance improvement of our model for anomaly detection tasks.

【 授权许可】

Unknown