Ensembles of Generative Adversarial Networks

@article{Wang2016EnsemblesOG,
  title={Ensembles of Generative Adversarial Networks},
  author={Yaxing Wang and Lichao Zhang and Joost van de Weijer},
  journal={CoRR},
  year={2016},
  volume={abs/1612.00991}
}
Ensembles are a popular way to improve results of discrimina tive CNNs. The combination of several networks trained starting from diff erent initializations improves results significantly. In this paper we investigate t he usage of ensembles of GANs. The specific nature of GANs opens up several new ways t o construct ensembles. The first one is based on the fact that in the minima x game which is played to optimize the GAN objective the generator network k eeps on changing even after the… CONTINUE READING
Recent Discussions
This paper has been referenced on Twitter 29 times over the past 90 days. VIEW TWEETS

Similar Papers

Loading similar papers…