Abstract:Semi-supervised anomaly detection identifies abnormal (testing) observations which are different from normal (training) observations. In many practical situations, anomalies are poorly insufficient and not well defined, while the normal data are easily sampled, have a wide variety, and may not be classified. For this paradigm, we propose a novel end-to-end deep network as an anomaly detector only trained on normal samples. Our architecture consists of a conditional variational auto-encoder (CVAE), a feature discriminator (FD), and an adversarially trained WGAN-GP discriminator. The CVAE is designed as a generator to reconstruct images. It leverages underlying category information and multivariate Gaussian distributions to regularize the latent space, enabling a smooth and informative manifold. For anomalies which have a certain similarity to normal data, we perform active negative training by generating potential outliers from the latent space to limit network generative capability. In order to capture data characteristics, we maximize the mutual information between the inputs and the latent codes by the FD. It enhances the relationship between the high-dimensional image space and corresponding encoded vectors. To promote reconstruction, a structural similarity loss is applied to robustly recover local texture details and the WGAN-GP discriminator is employed to aid in generating photo-realistic images. We distinguish anomalies by computing a reconstruction-based anomaly score. Different from recent encoder–decoder or GAN-based architectures, our approach considers input categories, constructs, and exploits a useful manifold in an unsupervised manner and has a stronger reconstruction capability. The experimental results demonstrate that the proposed approach outperforms state-of-the-art methods over several benchmark datasets.

Unsupervised Anomaly Detection Via Generative Adversarial Networks: Poster Abstract

DOPING: Generative Data Augmentation for Unsupervised Anomaly Detection with GAN

Anomaly Detection by Latent Regularized Dual Adversarial Networks.

Adversarially Learned Anomaly Detection

Supervised Anomaly Detection via Conditional Generative Adversarial Network and Ensemble Active Learning

GANomaly: Semi-supervised Anomaly Detection via Adversarial Training

Improving anomaly detection with a self-supervised task based on generative adversarial network

AGAD: Adversarial Generative Anomaly Detection

Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery

Anomaly Detection by Leveraging Incomplete Anomalous Knowledge with Anomaly-Aware Bidirectional GANs

Anomaly detection by using a combination of generative adversarial networks and convolutional autoencoders

ADGAN: A Scalable GAN-based Architecture for Image Anomaly Detection

A Survey on GANs for Anomaly Detection

on the effectiveness of generative adversarial network on anomaly detection

A Novel and Efficient CVAE-GAN-Based Approach With Informative Manifold for Semi-Supervised Anomaly Detection

GAN Ensemble for Anomaly Detection

Anomaly Detection by One Class Latent Regularized Networks

GAN-based anomaly detection: A review

GANAD: A GAN-based Method for Network Anomaly Detection

Generative adversarial nets for unsupervised outlier detection

Self-Supervised Attentive Generative Adversarial Networks for Video Anomaly Detection