COEGAN: Evaluating the Coevolution Effect in Generative Adversarial Networks

TL;DR

COEGAN introduces a coevolutionary approach to train GANs, automatically designing architectures and improving stability, effectively addressing issues like mode collapse on Fashion-MNIST and MNIST datasets.

Contribution

The paper presents COEGAN, a novel neuroevolution-based method that coevolves GAN components for stable training and automatic architecture design.

Findings

COEGAN discovers efficient architectures for Fashion-MNIST and MNIST.

COEGAN improves training stability and reduces mode collapse.

Compared to baseline and random search, COEGAN shows superior results.

Abstract

Generative adversarial networks (GAN) present state-of-the-art results in the generation of samples following the distribution of the input dataset. However, GANs are difficult to train, and several aspects of the model should be previously designed by hand. Neuroevolution is a well-known technique used to provide the automatic design of network architectures which was recently expanded to deep neural networks. COEGAN is a model that uses neuroevolution and coevolution in the GAN training algorithm to provide a more stable training method and the automatic design of neural network architectures. COEGAN makes use of the adversarial aspect of the GAN components to implement coevolutionary strategies in the training algorithm. Our proposal was evaluated in the Fashion-MNIST and MNIST dataset. We compare our results with a baseline based on DCGAN and also with results from a random search algorithm. We show that our method is able to discover efficient architectures in the Fashion-MNIST and MNIST datasets. The results also suggest that COEGAN can be used as a training algorithm for GANs to avoid common issues, such as the mode collapse problem.