Invited Talk
in
Workshop: Smooth Games Optimization and Machine Learning
An interpretation of GANs via online learning and game theory
Paulina Grnarova
Generative Adversarial Networks (GANs) have become one of the most powerful paradigms in learning real-world distributions. Despite this success, their minimax nature makes them fundamentally different to more classical generative models thus raising novel challenges; most notably in terms of training and evaluation. Indeed, finding a saddle-point is in general a harder task than converging to an extremum. We view the problem of training GANs as finding a mixed strategy in a zero-sum game. Building upon ideas from online learning and game theory, we propose (i) a novel training method with provable convergence to an equilibrium for semi-shallow GAN architectures, i.e. architectures where the discriminator is a one layer network and the generator is an arbitrary network and (ii) a natural metric for detecting non-convergence, namely the duality gap.
Live content is unavailable. Log in and register to view live content