Abstract:Optimal Transport (OT) based distances are powerful tools for machine learning to compare probability measures and manipulate them using OT maps. In this field, a setting of interest is semi-discrete OT, where the source measure $\mu$ is continuous, while the target $\nu$ is discrete. Recent works have shown that the minimax rate for the OT map is $\mathcal{O}(t^{-1/2})$ when using $t$ i.i.d. subsamples from each measure (two-sample setting). An open question is whether a better convergence rate can be achieved when the full information of the discrete measure $\nu$ is known (one-sample setting). In this work, we answer positively to this question by (i) proving an $\mathcal{O}(t^{-1})$ lower bound rate for the OT map, using the similarity between Laguerre cells estimation and density support estimation, and (ii) proposing a Stochastic Gradient Descent (SGD) algorithm with adaptive entropic regularization and averaging acceleration. To nearly achieve the desired fast rate, characteristic of non-regular parametric problems, we design an entropic regularization scheme decreasing with the number of samples. Another key step in our algorithm consists of using a projection step that permits to leverage the local strong convexity of the regularized OT problem. Our convergence analysis integrates online convex optimization and stochastic gradient techniques, complemented by the specificities of the OT semi-dual. Moreover, while being as computationally and memory efficient as vanilla SGD, our algorithm achieves the unusual fast rates of our theory in numerical experiments.

Entropic Approximation of -Optimal Transport Problems

Discrete Approximation of Optimal Transport on Compact Spaces

Approximation rates of entropic maps in semidiscrete optimal transport

A Sinkhorn-type Algorithm for Constrained Optimal Transport

Entropic estimation of optimal transport maps

Displacement smoothness of entropic optimal transport

Congestion and Penalization in Optimal Transport

Approximate Discrete Optimal Transport Plan with Auxiliary Measure Method.

Fast Entropic Regularized Optimal Transport Using Semidiscrete Cost Approximation

Neural Estimation Of Entropic Optimal Transport

Weak Optimal Entropy Transport Problems

Constrained Approximate Optimal Transport Maps

Approximating optimal transport with linear programs

On optimal transport of matrix-valued measures

Entropic Optimal Transport: Convergence of Potentials

Semi-Discrete Optimal Transport: Nearly Minimax Estimation With Stochastic Gradient Descent and Adaptive Entropic Regularization

Fast and Accurate Approximations of the Optimal Transport in Semi-Discrete and Discrete Settings

A semiconcavity approach to stability of entropic plans and exponential convergence of Sinkhorn's algorithm

The entropic optimal (self-)transport problem: Limit distributions for decreasing regularization with application to score function estimation