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

Banach Wasserstein GAN

2019-01-11
Jonas Adler, Sebastian Lunz

Abstract

Wasserstein Generative Adversarial Networks (WGANs) can be used to generate realistic samples from complicated image distributions. The Wasserstein metric used in WGANs is based on a notion of distance between individual images, which induces a notion of distance between probability distributions of images. So far the community has considered $\ell^2$ as the underlying distance. We generalize the theory of WGAN with gradient penalty to Banach spaces, allowing practitioners to select the features to emphasize in the generator. We further discuss the effect of some particular choices of underlying norms, focusing on Sobolev norms. Finally, we demonstrate a boost in performance for an appropriate choice of norm on CIFAR-10 and CelebA.

Abstract (translated by Google)
URL

https://arxiv.org/abs/1806.06621

PDF

https://arxiv.org/pdf/1806.06621


Similar Posts

Comments