Improved Techniques for Training GANs

We present a variety of new architectural features and training procedures that we apply to the generative adversarial networks (GANs) framework. Using our new techniques, we achieve state-of-the-art results in semi-supervised classification on MNIST, CIFAR-10 and SVHN. The generated images are of high quality as confirmed by a visual Turing test: Our model generates MNIST samples that humans cannot distinguish from real data, and CIFAR-10 samples that yield a human error rate of 21. 3%. We also present ImageNet samples with unprecedented resolution and show that our methods enable the model to learn recognizable features of ImageNet classes.

Authors

• Tim Salimans
• Ian Goodfellow Université de Montréal
• Wojciech Zaremba OpenAI
• Vicki Cheung OpenAI
• Alec Radford
• Xi Chen

Keywords

No keywords are indexed for this paper.