Skip to content

rdh1115/gans-and-wgans

 
 

Repository files navigation

A survey on Wasserstein GANs and vanilla GANs.

We first discuss the theory behind WGANs, most notably Optimal Transport. We then explore the mathematical theory behind GANs, introduce their limitations, and provide an overview of ways to address the training instability in practice. Finally, we talk about how the Wasserstein distance can be used as an alternative distribution measure, giving rise to WGAN.

We provide some experiments to compare the two on manually defined distributions here

See pdf for the full text.

About

Math 562 Winter term 2023 group project

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 93.8%
  • TeX 5.9%
  • Other 0.3%