Generative adversarial networks (GANs) have been effective for learning generative models for real-world data. However, accompanied with the generative tasks becoming more and more challenging, existing GANs (GAN and its variants) tend to suffer from different training problems such as instability and mode collapse. In this paper, we propose a novel GAN framework called evolutionary GANs (E-GANs) for stable GAN training and improved generative performance. Unlike existing GANs, which employ a predefined adversarial objective function alternately training a generator and a discriminator, we evolve a population of generators to play the adversarial game with the discriminator. Different adversarial training objectives are employed as mutation operations and each individual (i.e., generator candidature) are updated based on these mutations. Then, we devise 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 well-performing 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.

中文翻译：

---

进化生成对抗网络

生成对抗网络 (GAN) 在学习现实世界数据的生成模型方面非常有效。然而，随着生成任务变得越来越具有挑战性，现有的 GAN（GAN 及其变体）往往会遇到不同的训练问题，例如不稳定和模式崩溃。在本文中，我们提出了一种称为进化 GAN（E-GAN）的新型 GAN 框架，用于稳定的 GAN 训练和改进的生成性能。与现有的 GAN 不同，它采用预定义的对抗性目标函数交替训练生成器和判别器，我们进化了一组生成器来与判别器玩对抗游戏。不同的对抗训练目标被用作变异操作，并且每个个体（即生成器候选）基于这些变异进行更新。然后，我们设计了一种评估机制来衡量生成样本的质量和多样性，以便只保留性能良好的生成器并用于进一步训练。通过这种方式，E-GAN 克服了单个对抗性训练目标的局限性，并始终保留表现良好的后代，有助于 GAN 的进步和成功。在几个数据集上的实验表明，E-GAN 实现了令人信服的生成性能并减少了现有 GAN 固有的训练问题。E-GAN 克服了单个对抗性训练目标的局限性，并始终保留表现良好的后代，为 GAN 的进步和成功做出贡献。在几个数据集上的实验表明，E-GAN 实现了令人信服的生成性能并减少了现有 GAN 固有的训练问题。E-GAN 克服了单个对抗性训练目标的局限性，并始终保留表现良好的后代，为 GAN 的进步和成功做出贡献。在几个数据集上的实验表明，E-GAN 实现了令人信服的生成性能并减少了现有 GAN 固有的训练问题。