papers AI Learner
The Github is limit! Click to go to the new site.

Some Theoretical Properties of GANs

2018-03-21
G. Biau (LPSM), B. Cadre (ENS Rennes), M. Sangnier (LPSM), U. Tanielian (LPSM)

Abstract

Generative Adversarial Networks (GANs) are a class of generative algorithms that have been shown to produce state-of-the art samples, especially in the domain of image creation. The fundamental principle of GANs is to approximate the unknown distribution of a given data set by optimizing an objective function through an adversarial game between a family of generators and a family of discriminators. In this paper, we offer a better theoretical understanding of GANs by analyzing some of their mathematical and statistical properties. We study the deep connection between the adversarial principle underlying GANs and the Jensen-Shannon divergence, together with some optimality characteristics of the problem. An analysis of the role of the discriminator family via approximation arguments is also provided. In addition, taking a statistical point of view, we study the large sample properties of the estimated distribution and prove in particular a central limit theorem. Some of our results are illustrated with simulated examples.

Abstract (translated by Google)
URL

https://arxiv.org/abs/1803.07819

PDF

https://arxiv.org/pdf/1803.07819


Similar Posts

Comments