pith. sign in

arxiv: 1904.01184 · v1 · pith:ACG2JONWnew · submitted 2019-04-02 · 💻 cs.LG · stat.ML

Towards Efficient and Unbiased Implementation of Lipschitz Continuity in GANs

classification 💻 cs.LG stat.ML
keywords lipschitzcontinuitygradientnormalizationpenaltyspectralefficientgans
0
0 comments X
read the original abstract

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.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.