Dualing GANs

TL;DR

This paper introduces Dualing GANs, a novel approach that reformulates the GAN training process through duality to improve stability and provide an alternative training method, applicable to both linear and nonlinear discriminators.

Contribution

It proposes a dualization technique for GANs that stabilizes training and extends to nonlinear discriminators, offering a new perspective on GAN optimization.

Findings

Linear discriminator Dualing GANs remove training instability.

Extension to nonlinear discriminators offers an alternative training approach.

Demonstrates improved stability and convergence in GAN training.

Abstract

Generative adversarial nets (GANs) are a promising technique for modeling a distribution from samples. It is however well known that GAN training suffers from instability due to the nature of its maximin formulation. In this paper, we explore ways to tackle the instability problem by dualizing the discriminator. We start from linear discriminators in which case conjugate duality provides a mechanism to reformulate the saddle point objective into a maximization problem, such that both the generator and the discriminator of this 'dualing GAN' act in concert. We then demonstrate how to extend this intuition to non-linear formulations. For GANs with linear discriminators our approach is able to remove the instability in training, while for GANs with nonlinear discriminators our approach provides an alternative to the commonly used GAN training algorithm.