Bruno Lecouat, Chuan-Sheng Foo, Houssam Zenati, Vijay R. Chandrasekhar
TL;DR使用 GAN 进行流形正则化，采用 Monte Carlo 近似方法进行拉普拉斯规范化，结合 GAN 实现特征匹配 GAN，并在 CIFAR-10 数据集上实现了半监督学习并取得了最先进的结果。
Abstract
gans are powerful generative models that are able to model the manifold of natural images. We leverage this property to perform manifold regularization by approximating the →