Skip to yearly menu bar Skip to main content


Poster

Discriminator optimal transport

Akinori Tanaka

East Exhibition Hall B, C #116

Keywords: [ Generative Models ] [ Deep Learning ] [ Adversarial Networks ]


Abstract: Within a broad class of generative adversarial networks, we show that discriminator optimization process increases a lower bound of the dual cost function for the Wasserstein distance between the target distribution pp and the generator distribution pGpG. It implies that the trained discriminator can approximate optimal transport (OT) from pGpG to pp. Based on some experiments and a bit of OT theory, we propose discriminator optimal transport (DOT) scheme to improve generated images. We show that it improves inception score and FID calculated by un-conditional GAN trained by CIFAR-10, STL-10 and a public pre-trained model of conditional GAN trained by ImageNet.

Live content is unavailable. Log in and register to view live content