Y-GAN: Learning Dual Data Representations for Efficient Anomaly Detection

TL;DR

Y-GAN introduces a dual latent space auto-encoder for anomaly detection, separating semantic and residual features to improve detection accuracy across various datasets.

Contribution

It presents a novel Y-shaped auto-encoder with disentangled dual representations and a consistency loss, enhancing anomaly detection in a one-class learning setting.

Findings

Effective on multiple datasets including MNIST, FMNIST, CIFAR10, and PlantVillage.

Outperforms several recent anomaly detection models.

Provides efficient and robust anomaly detection through dual representation separation.

Abstract

We propose a novel reconstruction-based model for anomaly detection, called Y-GAN. The model consists of a Y-shaped auto-encoder and represents images in two separate latent spaces. The first captures meaningful image semantics, key for representing (normal) training data, whereas the second encodes low-level residual image characteristics. To ensure the dual representations encode mutually exclusive information, a disentanglement procedure is designed around a latent (proxy) classifier. Additionally, a novel consistency loss is proposed to prevent information leakage between the latent spaces. The model is trained in a one-class learning setting using normal training data only. Due to the separation of semantically-relevant and residual information, Y-GAN is able to derive informative data representations that allow for efficient anomaly detection across a diverse set of anomaly detection tasks. The model is evaluated in comprehensive experiments with several recent anomaly detection models using four popular datasets, i.e., MNIST, FMNIST and CIFAR10, and PlantVillage.