From Optimal Transport to Discrepancy

Authors Sebastian Neumayer, Gabriele Steidl

Publication date 4 February 2020

Abstract: A common way to quantify the ,,distance between measures is via their discrepancy, also known as maximum mean discrepancy (MMD). Discrepancies are related to Sinkhorn divergences $S_{v} a r e p s i l o n$ with appropriate cost functions as $v a r e p s i l o n o i n f t y$ . In the opposite direction, if $v a r e p s i l o n o 0$ , Sinkhorn divergences approach another important distance between measures, namely the Wasserstein distance or more generally optimal transport ,,distance. In this chapter, we investigate the limiting process for arbitrary measures on compact sets and Lipschitz continuous cost functions. In particular, we are interested in the behavior of the corresponding optimal potentials

h a t v a r p h i_{v} a r e p s i l o n

,

h a t p s i_{v} a r e p s i l o n

and

h a t v a r p h i_{K}

appearing in the dual formulation of the Sinkhorn divergences and discrepancies, respectively. While part of the results are known, we provide rigorous proofs for some relations which we have not found in this generality in the literature. Finally, we demonstrate the limiting process by numerical examples and show the behavior of the distances when used for the approximation of measures by point measures in a process called dithering.

This page was built for publication: From Optimal Transport to Discrepancy

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6334087)