# GAN Ensemble for Anomaly Detection

arXiv: https://arxiv.org/abs/2012.07988 · HTML: https://mlsanity.com/paper/2012.07988/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Dec 2020.

## Built on by (most cited first, 22 of 22)

- [Generalized Out-of-Distribution Detection: A Survey](https://mlsanity.com/paper/2110.11334/) (2021)
- [Trae Agent: An LLM-based Agent for Software Engineering with Test-time Scaling](https://mlsanity.com/paper/2507.23370/) (2025)
- [Omni-frequency Channel-selection Representations for Unsupervised Anomaly Detection](https://mlsanity.com/paper/2203.00259/) (2022)
- [A Survey on Ensemble Learning under the Era of Deep Learning](https://mlsanity.com/paper/2101.08387/) (2021)
- [Self-Supervised Masking for Unsupervised Anomaly Detection and Localization](https://mlsanity.com/paper/2205.06568/) (2022)
- [EMEF: Ensemble Multi-Exposure Image Fusion](https://mlsanity.com/paper/2305.12734/) (2023)
- [Anomaly Detection: How to Artificially Increase your F1-Score with a Biased Evaluation Protocol](https://mlsanity.com/paper/2106.16020/) (2021)
- [Scale-Aware Contrastive Reverse Distillation for Unsupervised Medical Anomaly Detection](https://mlsanity.com/paper/2503.13828/) (2025)
- [Double-Adversarial Activation Anomaly Detection: Adversarial Autoencoders are Anomaly Generators](https://mlsanity.com/paper/2101.04645/) (2021)
- [GraphAnoGAN: Detecting Anomalous Snapshots from Attributed Graphs](https://mlsanity.com/paper/2106.15504/) (2021)
- [Rethinking Multidimensional Discriminator Output for Generative Adversarial Networks](https://mlsanity.com/paper/2109.03378/) (2021)
- [Connective Reconstruction-based Novelty Detection](https://mlsanity.com/paper/2210.13917/) (2022)
- [Learning Invariant Rules from Data for Interpretable Anomaly Detection](https://mlsanity.com/paper/2211.13577/) (2022)
- [BenchReAD: A systematic benchmark for retinal anomaly detection](https://mlsanity.com/paper/2507.10492/) (2025)
- [CADE: Continual Weakly-supervised Video Anomaly Detection with Ensembles](https://mlsanity.com/paper/2512.06840/) (2025)
- [Detecting Irregular Network Activity with Adversarial Learning and Expert Feedback](https://mlsanity.com/paper/2210.02841/) (2022)
- [Anomaly Detection with Ensemble of Encoder and Decoder](https://mlsanity.com/paper/2303.06431/) (2023)
- [Additive Class Distinction Maps using Branched-GANs](https://mlsanity.com/paper/2305.02899/) (2023)
- [PARs: Predicate-based Association Rules for Efficient and Accurate Model-Agnostic Anomaly Explanation](https://mlsanity.com/paper/2312.10968/) (2023)
- [TemporalAugmenter: An Ensemble Recurrent Based Deep Learning Approach for Signal Classification](https://mlsanity.com/paper/2401.06970/) (2024)
- [Quantum enhanced ensemble GANs for anomaly detection in continuous biomanufacturing](https://mlsanity.com/paper/2508.21438/) (2025)
- [Automatic Unsupervised Ensemble Outlier Model Selection--Extended Version](https://mlsanity.com/paper/2605.16567/) (2026)

## Abstract (arXiv)

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.