Evolutionary Generative Adversarial Networks

TL;DR

E-GAN introduces an evolutionary framework for GAN training that uses multiple objectives and a selection mechanism to enhance stability and sample diversity, outperforming traditional GANs.

Contribution

The paper presents a novel evolutionary approach to GAN training, combining multiple objectives and selection to improve stability and generative quality.

Findings

E-GAN achieves better sample diversity and quality.

Reduces training instability and mode collapse.

Outperforms traditional GANs on several datasets.

Abstract

Generative adversarial networks (GAN) have been effective for learning generative models for real-world data. However, existing GANs (GAN and its variants) tend to suffer from training problems such as instability and mode collapse. In this paper, we propose a novel GAN framework called evolutionary generative adversarial networks (E-GAN) for stable GAN training and improved generative performance. Unlike existing GANs, which employ a pre-defined adversarial objective function alternately training a generator and a discriminator, we utilize different adversarial training objectives as mutation operations and evolve a population of generators to adapt to the environment (i.e., the discriminator). We also utilize an evaluation mechanism to measure the quality and diversity of generated samples, such that only well-performing generator(s) are preserved and used for further training. In this way, E-GAN overcomes the limitations of an individual adversarial training objective and always preserves the best offspring, contributing to progress in and the success of GANs. Experiments on several datasets demonstrate that E-GAN achieves convincing generative performance and reduces the training problems inherent in existing GANs.