ICML 2021spotlight3 citations

WGAN with an Infinitely Wide Generator Has No Spurious Stationary Points

Albert No, Taeho Yoon, Kwon Sehyun, Ernest K Ryu

Abstract

Generative adversarial networks (GAN) are a widely used class of deep generative models, but their minimax training dynamics are not understood very well. In this work, we show that GANs with a 2-layer infinite-width generator and a 2-layer finite-width discriminator trained with stochastic gradient ascent-descent have no spurious stationary points. We then show that when the width of the generator is finite but wide, there are no spurious stationary points within a ball whose radius becomes arbitrarily large (to cover the entire parameter space) as the width goes to infinity.

BibTeX
@InProceedings{pmlr-v139-no21a,
  title = 	 {WGAN with an Infinitely Wide Generator Has No Spurious Stationary Points},
  author =       {No, Albert and Yoon, Taeho and Sehyun, Kwon and Ryu, Ernest K},
  booktitle = 	 {Proceedings of the 38th International Conference on Machine Learning},
  pages = 	 {8205--8215},
  year = 	 {2021},
  editor = 	 {Meila, Marina and Zhang, Tong},
  volume = 	 {139},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {18--24 Jul},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v139/no21a/no21a.pdf},
  url = 	 {https://proceedings.mlr.press/v139/no21a.html},
  abstract = 	 {Generative adversarial networks (GAN) are a widely used class of deep generative models, but their minimax training dynamics are not understood very well. In this work, we show that GANs with a 2-layer infinite-width generator and a 2-layer finite-width discriminator trained with stochastic gradient ascent-descent have no spurious stationary points. We then show that when the width of the generator is finite but wide, there are no spurious stationary points within a ball whose radius becomes arbitrarily large (to cover the entire parameter space) as the width goes to infinity.}
}
WGAN with an Infinitely Wide Generator Has No Spurious Stationary Points · ICML 2021