Potential Flow Generator with L_2 Optimal Transport Regularity for Generative Models

08/29/2019

∙

by   Liu Yang, et al.

∙

2

∙

share

We propose a potential flow generator with L_2 optimal transport regularity, which can be easily integrated into a wide range of generative models including different versions of GANs and flow-based models. We show the correctness and robustness of the potential flow generator in several 2D problems, and illustrate the concept of "proximity" due to the L_2 optimal transport regularity. Subsequently, we demonstrate the effectiveness of the potential flow generator in image translation tasks with unpaired training data from the MNIST dataset and the CelebA dataset.