The statistical effect of entropic regularization in optimal transportation

06/09/2020
by   Eustasio del Barrio, et al.
0

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

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/19/2020

Entropy-regularized optimal transport on multivariate normal and q-normal distributions

Distance and divergence of the probability measures play a central role ...
research
01/06/2022

An Homogeneous Unbalanced Regularized Optimal Transport model with applications to Optimal Transport with Boundary

This work studies how the introduction of the entropic regularization te...
research
10/23/2018

Empirical Regularized Optimal Transport: Statistical Theory and Applications

We derive limit distributions for certain empirical regularized optimal ...
research
09/29/2022

Improving Generative Flow Networks with Path Regularization

Generative Flow Networks (GFlowNets) are recently proposed models for le...
research
01/24/2021

Entropy Partial Transport with Tree Metrics: Theory and Practice

Optimal transport (OT) theory provides powerful tools to compare probabi...
research
11/04/2021

Hard Negative Sampling via Regularized Optimal Transport for Contrastive Representation Learning

We study the problem of designing hard negative sampling distributions f...
research
10/17/2017

Smooth and Sparse Optimal Transport

Entropic regularization is quickly emerging as a new standard in optimal...

Please sign up or login with your details

Forgot password? Click here to reset