Robust Generative Adversarial Network

by   Shufei Zhang, et al.

Generative adversarial networks (GANs) are powerful generative models, but usually suffer from instability and generalization problem which may lead to poor generations. Most existing works focus on stabilizing the training of the discriminator while ignoring the generalization properties. In this work, we aim to improve the generalization capability of GANs by promoting the local robustness within the small neighborhood of the training samples. We also prove that the robustness in small neighborhood of training sets can lead to better generalization. Particularly, we design a robust optimization framework where the generator and discriminator compete with each other in a worst-case setting within a small Wasserstein ball. The generator tries to map the worst input distribution (rather than a Gaussian distribution used in most GANs) to the real data distribution, while the discriminator attempts to distinguish the real and fake distribution with the worst perturbation. We have proved that our robust method can obtain a tighter generalization upper bound than traditional GANs under mild assumptions, ensuring a theoretical superiority of RGAN over GANs. A series of experiments on CIFAR-10, STL-10 and CelebA datasets indicate that our proposed robust framework can improve on five baseline GAN models substantially and consistently.



There are no comments yet.


page 15


Unrolled Generative Adversarial Networks

We introduce a method to stabilize Generative Adversarial Networks (GANs...

On the Anomalous Generalization of GANs

Generative models, especially Generative Adversarial Networks (GANs), ha...

On the Discrimination-Generalization Tradeoff in GANs

Generative adversarial training can be generally understood as minimizin...

Robust GANs against Dishonest Adversaries

Robustness of deep learning models is a property that has recently gaine...

Robust Machine Learning via Privacy/Rate-Distortion Theory

Robust machine learning formulations have emerged to address the prevale...

Adversarial Learning with Local Coordinate Coding

Generative adversarial networks (GANs) aim to generate realistic data fr...

Lipizzaner: A System That Scales Robust Generative Adversarial Network Training

GANs are difficult to train due to convergence pathologies such as mode ...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.