Papers
Topics
Authors
Recent
Search
2000 character limit reached

Doubly Regularized Entropic Wasserstein Barycenters

Published 21 Mar 2023 in math.OC, cs.LG, and stat.ML | (2303.11844v1)

Abstract: We study a general formulation of regularized Wasserstein barycenters that enjoys favorable regularity, approximation, stability and (grid-free) optimization properties. This barycenter is defined as the unique probability measure that minimizes the sum of entropic optimal transport (EOT) costs with respect to a family of given probability measures, plus an entropy term. We denote it (λ,τ)(\lambda,\tau)-barycenter, where λ\lambda is the inner regularization strength and τ\tau the outer one. This formulation recovers several previously proposed EOT barycenters for various choices of λ,τ≥0\lambda,\tau \geq 0 and generalizes them. First, in spite of -- and in fact owing to -- being \emph{doubly} regularized, we show that our formulation is debiased for τ=λ/2\tau=\lambda/2: the suboptimality in the (unregularized) Wasserstein barycenter objective is, for smooth densities, of the order of the strength λ2\lambda^2 of entropic regularization, instead of max⁡{λ,τ}\max\{\lambda,\tau\} in general. We discuss this phenomenon for isotropic Gaussians where all (λ,τ)(\lambda,\tau)-barycenters have closed form. Second, we show that for λ,τ>0\lambda,\tau>0, this barycenter has a smooth density and is strongly stable under perturbation of the marginals. In particular, it can be estimated efficiently: given nn samples from each of the probability measures, it converges in relative entropy to the population barycenter at a rate n−1/2n^{-1/2}. And finally, this formulation lends itself naturally to a grid-free optimization algorithm: we propose a simple \emph{noisy particle gradient descent} which, in the mean-field limit, converges globally at an exponential rate to the barycenter.

Authors (1)
Citations (27)

Summary

No one has generated a summary of this paper yet.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.

Tweets

Sign up for free to view the 1 tweet with 1 like about this paper.