Link to our paper - https://arxiv.org/abs/1705.07215
Pytorch implementation (thanks!) - https://github.com/jfsantos/dragan-pytorch
Procedure (to use our algorithm):
Interesting discussion with Ian Goodfellow and Martin Arjovsky on why GANs are unstable and where improvements come from
https://www.facebook.com/kodali.naveen.90/posts/1047257878740881
An interesting new paper by Fedus et.al came out following this (Many paths to equilibria)
https://arxiv.org/abs/1710.08446
Some of the repositories that would be helpful and which helped us (big thanks!):
https://github.com/igul222/improved_wgan_training
https://github.com/wiseodd/generative-models/tree/master/GAN
https://github.com/openai/improved-gan/tree/master/inception_score