DeepAI AI Chat
Log In Sign Up

Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields

by   Thomas Unterthiner, et al.
Johannes Kepler University Linz

Generative adversarial networks (GANs) evolved into one of the most successful unsupervised techniques for generating realistic images. Even though it has recently been shown that GAN training converges, GAN models often end up in local Nash equilibria that are associated with mode collapse or otherwise fail to model the target distribution. We introduce Coulomb GANs, which pose the GAN learning problem as a potential field of charged particles, where generated samples are attracted to training set samples but repel each other. The discriminator learns a potential field while the generator decreases the energy by moving its samples along the vector (force) field determined by the gradient of the potential field. Through decreasing the energy, the GAN model learns to generate samples according to the whole target distribution and does not only cover some of its modes. We prove that Coulomb GANs possess only one Nash equilibrium which is optimal in the sense that the model distribution equals the target distribution. We show the efficacy of Coulomb GANs on a variety of image datasets. On LSUN and celebA, Coulomb GANs set a new state of the art and produce a previously unseen variety of different samples.


page 8

page 9

page 18

page 19


GANs May Have No Nash Equilibria

Generative adversarial networks (GANs) represent a zero-sum game between...

Seeing What a GAN Cannot Generate

Despite the success of Generative Adversarial Networks (GANs), mode coll...

On GANs and GMMs

A longstanding problem in machine learning is to find unsupervised metho...

Beyond Local Nash Equilibria for Adversarial Networks

Save for some special cases, current training methods for Generative Adv...

Characterizing GAN Convergence Through Proximal Duality Gap

Despite the accomplishments of Generative Adversarial Networks (GANs) in...

Effects of Dataset properties on the training of GANs

Generative Adversarial Networks are a new family of generative models, f...

On Convergence and Stability of GANs

We propose studying GAN training dynamics as regret minimization, which ...