Entropic regularization of Monge's problem

TL;DR

This paper analyzes the limit of entropically regularized optimal transport as regularization vanishes, providing a variational framework, explicit convergence results, and a second-order expansion of the entropic transport cost.

Contribution

It resolves the entropic selection problem and derives an explicit second-order expansion, revealing the interplay between entropy and transport cost in the zero-regularization limit.

Findings

EOT minimizers converge to a distinguished optimal transport plan

Explicit second-order expansion of the entropic transport cost

Sharp characterization of the asymptotic tradeoff between entropy and cost

Abstract

We study the vanishing-regularization limit of entropically regularized optimal transport (EOT) for the Euclidean distance cost $c(x,y)=\|x-y\|$ in dimension $d>1$. We develop a comprehensive variational convergence framework that entails two main results. First, we resolve the longstanding entropic selection problem: the EOT minimizer converges to a distinguished optimal transport plan that is characterized explicitly as the solution of a constrained EOT problem on each transport ray. Denoting by $\varepsilon>0$ the regularization parameter, this selection holds for all $o(\varepsilon)$-approximate minimizers, with sharp failure at the $O(\varepsilon)$ scale. Second, we establish an explicit second-order expansion of the entropic transport cost. The second-order term encodes the geometry of the regularization and reveals the optimal asymptotic tradeoff between entropy and transport cost.