Beyond Local Nash Equilibria for Adversarial Networks

06/18/2018
by   Frans A. Oliehoek, et al.
4

Save for some special cases, current training methods for Generative Adversarial Networks (GANs) are at best guaranteed to converge to a `local Nash equilibrium` (LNE). Such LNEs, however, can be arbitrarily far from an actual Nash equilibrium (NE), which implies that there are no guarantees on the quality of the found generator or classifier. This paper proposes to model GANs explicitly as finite games in mixed strategies, thereby ensuring that every LNE is an NE. With this formulation, we propose a solution method that is proven to monotonically converge to a resource-bounded Nash equilibrium (RB-NE): by increasing computational resources we can find better solutions. We empirically demonstrate that our method is less prone to typical GAN problems such as mode collapse, and produces solutions that are less exploitable than those produced by GANs and MGANs, and closely resemble theoretical predictions about NEs.

READ FULL TEXT

page 7

page 8

page 14

page 17

page 19

page 21

research
12/02/2017

GANGs: Generative Adversarial Network Games

Generative Adversarial Networks (GAN) have become one of the most succes...
research
02/21/2020

GANs May Have No Nash Equilibria

Generative adversarial networks (GANs) represent a zero-sum game between...
research
06/11/2019

A Closer Look at the Optimization Landscapes of Generative Adversarial Networks

Generative adversarial networks have been very successful in generative ...
research
03/14/2022

On the Nash equilibrium of moment-matching GANs for stationary Gaussian processes

Generative Adversarial Networks (GANs) learn an implicit generative mode...
research
08/29/2017

Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields

Generative adversarial networks (GANs) evolved into one of the most succ...
research
05/11/2021

Characterizing GAN Convergence Through Proximal Duality Gap

Despite the accomplishments of Generative Adversarial Networks (GANs) in...
research
10/23/2018

Finding Mixed Nash Equilibria of Generative Adversarial Networks

We reconsider the training objective of Generative Adversarial Networks ...

Please sign up or login with your details

Forgot password? Click here to reset