Abstract

Generative Adversarial Networks (GANs) have breathed new life into research on generative models. Generative models promise to be able to learn rich structural representations from unsupervised data, enabling data-efficient modelling in complex domains. The talk is divided into three parts.

• The first part introduces the basic GAN approach, understanding it both on the statistical level in terms of minimizing a divergence between probability distributions and algorithmically in terms of a smooth two-player game.
• The second part discusses problems in the GAN approach and consolidates recent research by highlighting problems both in the statistical viewpoint (existence of divergences) and in the algorithmic viewpoint (convergence of the GAN game), making recommendations for practical use of GAN models.
• The third part discusses the relationship to other generative modelling approaches, potential applications of GANs and GAN -type approximations, and raises open problems for future research.

Speaker: Sebastian Nowozin, Principal Researcher, Machine Intelligence and Perception group, Microsoft Research, Cambridge, UK

Bio: Sebastian Nowozin is a machine learning researcher and manager of the Machine Intelligence and Perception group at Microsoft Research Cambridge, UK. He completed his PhD in 2009 at the Max Planck Institute for Biological Cybernetics in Tübingen, Germany. His research is in the area of probabilistic models, deep learning, and applications to computer vision problems. His research has received awards including best paper prizes at CVPR and the pattern recognition award from the German Pattern Recognition Society. At Microsoft his research and code has been shipped in Xbox, Azure ML, and Hololens.