Some Theoretical Properties of GANs - Sorbonne Université Accéder directement au contenu
Article Dans Une Revue Annals of Statistics Année : 2020

Some Theoretical Properties of GANs

Résumé

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.
Fichier principal
Vignette du fichier
adversarial.pdf (786.05 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01737975 , version 1 (20-03-2018)

Identifiants

Citer

Gérard Biau, Benoît Cadre, M. Sangnier, U. Tanielian. Some Theoretical Properties of GANs. Annals of Statistics, 2020, 48 (3), pp.1539-1566. ⟨10.1214/19-AOS1858⟩. ⟨hal-01737975⟩
256 Consultations
286 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More