Generative adversarial networks (GANs) are a technique for learning generative models of complex data distributions from samples. Despite remarkable advances in generating realistic images, a major shortcoming of GANs is the fact that they tend to produce samples with little diversity, even when trained on diverse datasets. This phenomenon, known as mode collapse, has been the focus of much recent work. We study a principled approach to handling mode collapse, which we call packing. The main idea is to modify the discriminator to make decisions based on multiple samples from the same class, either real or artificially generated. We draw analysis tools from binary hypothesis testing---in particular the seminal result of Blackwell---to prove a fundamental connection between packing and mode collapse. We show that packing naturally penalizes generators with mode collapse, thereby favoring generator distributions with less mode collapse during the training process. Numerical experiments on benchmark datasets suggest that packing provides significant improvements.

Their performance is well measured by the cross-entropy (CE) of the model distribution relative to the data distribution.

Neural Information Processing Systems http://nips.cc/

Exten-sive experiments show the effectiveness of this measure to rank different GANmodels and capture the typical GAN failure scenarios, including mode collapse andnon-convergentbehaviours.

Neural Information Processing Systems http://nips.cc/

Toevaluate theseGANs,weusedagroupofrealdatasets to set-up a benchmarking system and implemented three of the most recent techniques. For comparison purposes, we created two baseline methodsusingBayesiannetworks.