Can I read DO-GAN: A Double Oracle Framework for Generative Adversarial Networks on EtoBox?

DO-GAN: A Double Oracle Framework for Generative Adversarial Networks by Aung, Aye Phyu Phyu; Wang, Xinrun; Yu, Runsheng; An, Bo; Jayavelu, Senthilnath; Li, Xiaoli is a scholarly article available to read on EtoBox.

What is DO-GAN: A Double Oracle Framework for Generative Adversarial Networks about?

In this paper, we propose a new approach to train Generative Adversarial Networks (GANs) 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 players' strategies as the trained models of generator and discriminator from the best response oracles. We then compute the meta-strategies using a linear program. For scalability of the framework where multiple generators and discriminator best responses are stored in the memory, we propose two solutions: 1) pruning the weakly-dominated players' strategies to keep the oracles from becoming intractable; 2) applying continual learning to retain the previous knowledge of the networks. We apply our framework to established GAN architectures such as vanilla GAN, Deep Convolutional GAN, Spectral Normalization GAN and Stacked GAN. Finally, we conduct experiments on MNIST, CIF

Author

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

Published

2021

Language

EN