GANGs: Generative Adversarial Network Games

Abstract

Generative Adversarial Networks (GAN) have become one of the most successful frameworks for unsupervised generative modeling. As GANs are difficult to train much research has focused on this. However, very little of this research has directly exploited game-theoretic techniques. We introduce Generative Adversarial Network Games (GANGs), which explicitly model a finite zero-sum game between a generator ($G$) and classifier ($C$) that use mixed strategies. The size of these games precludes exact solution methods, therefore we define resource-bounded best responses (RBBRs), and a resource-bounded Nash Equilibrium (RB-NE) as a pair of mixed strategies such that neither $G$ or $C$ can find a better RBBR. The RB-NE solution concept is richer than the notion of `local Nash equilibria' in that it captures not only failures of escaping local optima of gradient descent, but applies to any approximate best response computations, including methods with random restarts. To validate our approach, we solve GANGs with the Parallel Nash Memory algorithm, which provably monotonically converges to an RB-NE. We compare our results to standard GAN setups, and demonstrate that our method deals well with typical GAN problems such as mode collapse, partial mode coverage and forgetting.

9 pages, 5 figures

References in corpus (9)

• Improved Training of Wasserstein GANs
• Improved Techniques for Training GANs
• Energy-based Generative Adversarial Network
• f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization
• Online Learning with Predictable Sequences
• Optimization, Learning, and Games with Predictable Sequences
• Do GANs actually learn the distribution? An empirical study
• Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields
• Efficient Regret Minimization in Non-Convex Games

Cited by in corpus (8)

• Stabilizing Generative Adversarial Networks: A Survey
• Continuous sign language recognition from wearable IMUs using deep capsule networks and game theory
• From Poincaré Recurrence to Convergence in Imperfect Information Games: Finding Equilibrium via Regularization
• A game-theoretic approach for Generative Adversarial Networks
• AlphaGAN: Fully Differentiable Architecture Search for Generative Adversarial Networks
• Semi-supervised Conditional GAN for Simultaneous Generation and Detection of Phishing URLs: A Game theoretic Perspective
• From open learners to open games
• Generative Adversarial Networks and Adversarial Autoencoders: Tutorial and Survey