Generative Adversarial Learning of Sinkhorn Algorithm Initializations

The Sinkhorn algorithm is the state-of-the-art to approximate solutions of entropic optimal transport (OT) distances between discrete probability distributions. We show that meticulously training a neural network to learn initializations to the algorithm via the entropic OT dual problem can significantly speed up convergence, while maintaining desirable properties of the Sinkhorn algorithm, such as differentiability and parallelizability. We train our predictive network in an adversarial fashion using a second, generating network and a self-supervised bootstrapping loss. The predictive network is …

adversarial adversarial learning algorithm art convergence cs.lg generative learn math.oc network neural network probability show solutions speed state stat.ml the algorithm training transport via