no code implementations • 13 Sep 2021 • Yao Chen, Qingyi Gao, Xiao Wang
The Wasserstein GAN (WGAN) leverages the Wasserstein distance to avoid the caveats in the minmax two-player training of GANs but has other defects such as mode collapse and lack of metric to detect the convergence.
no code implementations • 25 Sep 2019 • Yao Chen, Qingyi Gao, Xiao Wang
We further provide a rigorous probabilistic interpretation of our model under the framework of maximum likelihood estimation.
no code implementations • 25 Sep 2019 • Hao Chen, Zhanfeng Mo, Qingyi Gao, Zhouwang Yang, Xiao Wang
To better understand the unsupervised model, GANs, we establish the generalization bound, which uniformly holds with respect to the choice of generators.