Towards Efficient and Unbiased Implementation of Lipschitz Continuity in GANs

2 Apr 2019  ·  Zhiming Zhou, Jian Shen, Yuxuan Song, Wei-Nan Zhang, Yong Yu ·

Lipschitz continuity recently becomes popular in generative adversarial networks (GANs). It was observed that the Lipschitz regularized discriminator leads to improved training stability and sample quality. The mainstream implementations of Lipschitz continuity include gradient penalty and spectral normalization. In this paper, we demonstrate that gradient penalty introduces undesired bias, while spectral normalization might be over restrictive. We accordingly propose a new method which is efficient and unbiased. Our experiments verify our analysis and show that the proposed method is able to achieve successful training in various situations where gradient penalty and spectral normalization fail.

PDF Abstract


  Add Datasets introduced or used in this paper

Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.