The statistical effect of entropic regularization in optimal\n transportation

2020/06/09 by Eustasio del Barrio, del Barrio, Eustasio, Jean–Michel Loubes +1 · 2 citations
Mathematics · Physics and Astronomy · #Advanced Thermodynamics and Statistical Mechanics #FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Markov Chains and Monte Carlo Methods #Optimization and Control (math.OC) #Statistical Mechanics and Entropy #Statistics Theory (math.ST)

paper · pdf · doi:10.48550/arxiv.2006.05199

openalex publication_date 2020/06/09 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28

Abstract

We propose to tackle the problem of understanding the effect of\nregularization in Sinkhorn algotihms. In the case of Gaussian distributions we\nprovide a closed form for the regularized optimal transport which enables to\nprovide a better understanding of the effect of the regularization from a\nstatistical framework.\n

Citations

• Stochastic Optimization for Large-scale Optimal Transport
• Computational Optimal Transport: Complexity by Accelerated Gradient Descent Is Better Than by Sinkhorn's Algorithm

Cited by

• Conditional Wasserstein Barycenters and Interpolation/Extrapolation of Distributions
• On the Global Optimality of Linear Policies for Sinkhorn Distributionally Robust Linear Quadratic Control

Related

• The entropic optimal (self-)transport problem: Limit distributions for decreasing regularization with application to score function estimation
• Empirical Regularized Optimal Transport: Statistical Theory and\n Applications
• Central limit theorems for entropy-regularized optimal transport on finite spaces and statistical applications
• Beyond entropic regularization: Debiased Gaussian estimators for discrete optimal transport and general linear programs
• Weak limits of entropy regularized Optimal Transport; potentials, plans and divergences