Corpus ID: 227334945

Sobolev Wasserstein GAN

@article{Xu2020SobolevWG,
  title={Sobolev Wasserstein GAN},
  author={Minkai Xu and Zhiming Zhou and Guansong Lu and J. Tang and W. Zhang and Y. Yu},
  journal={ArXiv},
  year={2020},
  volume={abs/2012.03420}
}
Wasserstein GANs (WGANs), built upon the Kantorovich-Rubinstein (KR) duality of Wasserstein distance, is one of the most theoretically sound GAN models. However, in practice it does not always outperform other variants of GANs. This is mostly due to the imperfect implementation of the Lipschitz condition required by the KR duality. Extensive work has been done in the community with different implementations of the Lipschitz constraint, which, however, is still hard to satisfy the restriction… Expand

References

SHOWING 1-10 OF 41 REFERENCES
Lipschitz Generative Adversarial Nets
  • 34
  • PDF
On the regularization of Wasserstein GANs
  • 103
  • PDF
Gradient descent GAN optimization is locally stable
  • 224
  • PDF
Banach Wasserstein GAN
  • 57
  • PDF
The Cramer Distance as a Solution to Biased Wasserstein Gradients
  • 186
  • Highly Influential
  • PDF
Which Training Methods for GANs do actually Converge?
  • 512
  • Highly Influential
  • PDF
Improved Training of Wasserstein GANs
  • 4,218
  • Highly Influential
  • PDF
GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
  • 2,693
  • PDF
A Convex Duality Framework for GANs
  • 36
  • PDF
On Convergence and Stability of GANs
  • 265
...
1
2
3
4
5
...