Wasserstein GAN可以执行PCA

Feb, 2019

Wasserstein GAN Can Perform PCA

Jaewoong Cho, Changho Suh

TL;DR研究生成敌对网络是一个强大的框架，而文中的研究将GANs应用于简单的线性生成器高斯数据场景下，发现原GAN无法恢复最优PCA解，而Wasserstein GAN可以在样本大小的极限下接近PCA解，因此可能成为一种基于广泛数据设置的最优GAN体系结构的基础。

Abstract

generative adversarial networks (gans) have become a powerful framework to learn generative models that arise across a wide variety of domains. While there has been a recent surge in the development of numerous G