DO-GAN: A Double Oracle Framework for Generative Adversarial Networks

Aye Phyu Phyu Aung, Xinrun Wang, Runsheng Yu, Bo An, Senthilnath Jayavelu, Xiaoli Li

Keywords: GAN, Generative Models, Adversarial Networks, Game Theory

Abstract: In this paper, we propose a new approach to train Generative Adversarial Networks (GAN) where we deploy a double-oracle framework using the generator and discriminator oracles. GAN is essentially a two-player zero-sum game between the generator and the discriminator. Training GANs is challenging as a pure Nash equilibrium may not exist and even finding the mixed Nash equilibrium is difficult as GANs have a large-scale strategy space. In DO-GAN, we extend the double oracle framework to GANs. We first generalize the player strategies as the trained models of generator and discriminator from the best response oracles. We then compute the meta-strategies using a linear program. Next, we prune the weakly-dominated player strategies to keep the oracles from becoming intractable. We apply our framework to established architectures such as vanilla GAN, Deep Convolutional GAN, Spectral Normalization GAN and Stacked GAN. Finally, we conduct evaluations on MNIST, CIFAR-10 and CelebA datasets and show that DO-GAN variants have significant improvements in both subjective qualitative evaluation and quantitative metrics, compared with their respective GAN architectures.

One-sentence Summary: We deploy a double-oracle framework to GAN architectures using the generator and discriminator oracles which results in significant improvements over the adapted models.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Reviewed Version (pdf): https://openreview.net/references/pdf?id=iAUCQ2Hmtx