Novelty Detection via Robust Variational Autoencoding

Lai, Chieh-Hsin; Zou, Dongmian

doi:10.48550/arxiv.2006.05534

Cited by 1 publication

(1 citation statement)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thanks to the marvelous representation ability of DNNs, many reconstruction-based methods using DNNs are developed for UAD in recent years. They mainly employ deep generative models (Zenati et al, 2018;Perera et al, 2019;Lai et al, 2020b) or autoencoders (AE) (Chen et al, 2017;Pidhorskyi et al, 2018;Abati et al, 2019) to reconstruct data and determine abnormality of data via its reconstruction error. For instance, DAGMM (Zong et al, 2018) feed the latent representations of the AE into a gaussian mixture model and jointly optimize them.…”

Section: Related Workmentioning

confidence: 99%

SLA$^2$P: Self-supervised Anomaly Detection with Adversarial Perturbation

Wang¹,

Qin²,

Wei³

et al. 2021

Preprint

View full text Add to dashboard Cite

Anomaly detection is a fundamental yet challenging problem in machine learning due to the lack of label information. In this work, we propose a novel and powerful framework, dubbed as SLA 2 P, for unsupervised anomaly detection. After extracting representative embeddings from raw data, we apply random projections to the features and regard features transformed by different projections as belonging to distinct pseudo classes. We then train a classifier network on these transformed features to perform self-supervised learning. Next we add adversarial perturbation to the transformed features to decrease their softmax scores of the predicted labels and design anomaly scores based on the predictive uncertainties of the classifier on these perturbed features.Our motivation is that because of the relatively small number and the decentralized modes of anomalies, 1) the pseudo label classifier's training concentrates more on learning the semantic information of normal data rather than anomalous data; 2) the transformed features of the normal data are more robust to the perturbations than those of the anomalies. Consequently, the perturbed transformed features of anomalies fail to be classified well and accordingly have lower anomaly scores than those of the normal samples. Extensive experiments on image, text and inherently tabular benchmark datasets back up our findings and indicate that SLA 2 P achieves state-of-the-art results on unsupervised anomaly detection tasks consistently.

show abstract