Entropic regularization of continuous optimal transport problems

Christian Clason, Dirk A. Lorenz, Hinrich Mahler, Benedikt Wirth

Published 2019-06-04Version 1

We analyze continuous optimal transport problems in the so-called Kantorovich form, where we seek a transport plan between two marginals that are probability measures on compact subsets of Euclidean space. We consider the case of regularization with the negative entropy, which has attracted attention because it can be solved in the discrete case using the very simple Sinkhorn algorithm. We first analyze the problem in the context of classical Fenchel duality and derive a strong duality result for a predual problem in the space of continuous functions. However, this problem may not admit a minimizer, which prevents obtaining primal-dual optimality conditions that can be used to justify the Sinkhorn algorithm on the continuous level. We then show that the primal problem is naturally analyzed in the Orlicz space of functions with finite entropy and derive a dual problem in the corresponding dual space, for which existence can be shown and primal-dual optimality conditions can be derived. For marginals that do not have finite entropy, we finally show Gamma-convergence of the regularized problem with smoothed marginals to the original Kantorovich problem.