The statistical effect of entropic regularization in optimal transportation
arXiv:2006.05199v224 citations
AI Analysis
This provides theoretical insights into regularization in optimal transport, though it is incremental as it focuses on a specific distribution case.
The authors investigated how entropic regularization affects optimal transport algorithms, specifically deriving a closed-form solution for Gaussian distributions to statistically analyze regularization effects.
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.