Training bidirectional generative adversarial networks with hints

作者：

Highlights：

• The BiGAN has an encoder, in addition to the generator and discriminator of GAN.

• This encoder coupled with the generator allows defining extra loss terms as hints.

• We experiment on five image data sets, MNIST, UT-Zap50K, GTSRB, Cifar10, and CelebA.

• With these different hints, BiGAN generates higher quality and more diverse images.

摘要

•The BiGAN has an encoder, in addition to the generator and discriminator of GAN.•This encoder coupled with the generator allows defining extra loss terms as hints.•We experiment on five image data sets, MNIST, UT-Zap50K, GTSRB, Cifar10, and CelebA.•With these different hints, BiGAN generates higher quality and more diverse images.

论文关键词：Generative Modeling,Generative Adversarial Networks,Unsupervised Learning,Autoencoders,Neural Networks,Deep Learning

论文评审过程：Received 23 May 2019, Revised 15 January 2020, Accepted 28 February 2020, Available online 29 February 2020, Version of Record 5 March 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107320