# Ensembles of Generative Adversarial Networks

arXiv: https://arxiv.org/abs/1612.00991 · HTML: https://mlsanity.com/paper/1612.00991/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Dec 2016.

## Built on by (most cited first, 8 of 8)

- [MG-GAN: A Multi-Generator Model Preventing Out-of-Distribution Samples in Pedestrian Trajectory Prediction](https://mlsanity.com/paper/2108.09274/) (2021)
- [RareGAN: Generating Samples for Rare Classes](https://mlsanity.com/paper/2203.10674/) (2022)
- [Ensembles of GANs for synthetic training data generation](https://mlsanity.com/paper/2104.11797/) (2021)
- [A Realism Metric for Generated LiDAR Point Clouds](https://mlsanity.com/paper/2208.14958/) (2022)
- [EFFGAN: Ensembles of fine-tuned federated GANs](https://mlsanity.com/paper/2206.11682/) (2022)
- [A Bias-Variance Decomposition for Ensembles over Multiple Synthetic Datasets](https://mlsanity.com/paper/2402.03985/) (2024)
- [Beyond a Single Mode: GAN Ensembles for Diverse Medical Data Generation](https://mlsanity.com/paper/2503.24258/) (2025)
- [A New Paradigm for Generative Adversarial Networks based on Randomized Decision Rules](https://mlsanity.com/paper/2306.13641/) (2023)

## Abstract (arXiv)

Ensembles are a popular way to improve results of discriminative CNNs. The combination of several networks trained starting from different initializations improves results significantly. In this paper we investigate the usage of ensembles of GANs. The specific nature of GANs opens up several new ways to construct ensembles. The first one is based on the fact that in the minimax game which is played to optimize the GAN objective the generator network keeps on changing even after the network can be considered optimal. As such ensembles of GANs can be constructed based on the same network initialization but just taking models which have different amount of iterations. These so-called self ensembles are much faster to train than traditional ensembles. The second method, called cascade GANs, redirects part of the training data which is badly modeled by the first GAN to another GAN. In experiments on the CIFAR10 dataset we show that ensembles of GANs obtain model probability distributions which better model the data distribution. In addition, we show that these improved results can be obtained at little additional computational cost.