GAN Ensemble for Anomaly Detection
This work addresses the problem of improving anomaly detection performance for practitioners using GAN-based methods, representing an incremental improvement over existing GAN approaches.
This paper proposes using GAN ensembles for anomaly detection, motivated by their superior performance in generation tasks. The ensemble, where multiple generators and discriminators are trained together, better models normal data distributions, leading to improved anomaly detection compared to single GANs.
When formulated as an unsupervised learning problem, anomaly detection often requires a model to learn the distribution of normal data. Previous works apply Generative Adversarial Networks (GANs) to anomaly detection tasks and show good performances from these models. Motivated by the observation that GAN ensembles often outperform single GANs in generation tasks, we propose to construct GAN ensembles for anomaly detection. In the proposed method, a group of generators and a group of discriminators are trained together, so every generator gets feedback from multiple discriminators, and vice versa. Compared to a single GAN, a GAN ensemble can better model the distribution of normal data and thus better detect anomalies. Our theoretical analysis of GANs and GAN ensembles explains the role of a GAN discriminator in anomaly detection. In the empirical study, we evaluate ensembles constructed from four types of base models, and the results show that these ensembles clearly outperform single models in a series of tasks of anomaly detection.