Beyond Local Nash Equilibria for Adversarial Networks



Oliehoek, Frans A ORCID: 0000-0003-4372-5055, Savani, Rahul ORCID: 0000-0003-1262-7831, Gallego, Jose, van der Pol, Elise and Gross, Roderich
(2019) Beyond Local Nash Equilibria for Adversarial Networks. Springer International Publishing.

[img] Text
1806.07268v1.pdf - Submitted version

Download (2MB)

Abstract

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.

Item Type: Other
Additional Information: Supersedes arXiv:1712.00679; v2 includes Fictitious GAN in the related work and refers to Danskin (1981)
Uncontrolled Keywords: cs.LG, cs.LG, cs.GT, stat.ML
Depositing User: Symplectic Admin
Date Deposited: 27 Jul 2018 14:07
Last Modified: 19 Jan 2023 01:30
DOI: 10.1007/978-3-030-31978-6_7
Related URLs:
URI: https://livrepository.liverpool.ac.uk/id/eprint/3024066