BriefGPT.xyz
Sep, 2017
关于Wasserstein GAN正则化的研究
On the regularization of Wasserstein GANs
HTML
PDF
Henning Petzka, Asja Fischer, Denis Lukovnicov
TL;DR
本文提出了一种使用正则化项的生成对抗网络（GANs）的训练方法，以加强Lipschitz限制约束，该方法通过实验数据验证其有效性。
Abstract
Since their invention,
generative adversarial networks
(GANs) have become a popular approach for learning to model a distribution of real (unlabeled) data. Convergence problems during training are overcome by
wasserstei
→