Generative Adversarial Parallelization
This addresses training challenges in GANs for unsupervised learning, offering an incremental improvement in mode coverage and convergence.
The paper tackles the problem of training Generative Adversarial Networks (GANs), which are difficult to train and tend to ignore modes of the data distribution, by proposing Generative Adversarial Parallelization (GAP), a framework where multiple GANs are trained simultaneously with shared discriminators, leading to improved convergence and mode coverage.
Generative Adversarial Networks have become one of the most studied frameworks for unsupervised learning due to their intuitive formulation. They have also been shown to be capable of generating convincing examples in limited domains, such as low-resolution images. However, they still prove difficult to train in practice and tend to ignore modes of the data generating distribution. Quantitatively capturing effects such as mode coverage and more generally the quality of the generative model still remain elusive. We propose Generative Adversarial Parallelization, a framework in which many GANs or their variants are trained simultaneously, exchanging their discriminators. This eliminates the tight coupling between a generator and discriminator, leading to improved convergence and improved coverage of modes. We also propose an improved variant of the recently proposed Generative Adversarial Metric and show how it can score individual GANs or their collections under the GAP model.