Skip to content

Generative Adverserial Networks

Have two networks: a generative network $G(z)zD(Y)$.

See also:

Background

Mode collapse

A commonly encountered failure case for GANs where the generator learns to produce samples with extremely low variety.

Deconvolutions

Example architectures

DCGAN

  • https://github.com/Newmu/dcgan_code (theano)
  • https://github.com/carpedm20/DCGAN-tensorflow

pix2pix

WassersteinGAN

  • https://github.com/martinarjovsky/WassersteinGAN

Bibliography