Fast Computation of Wasserstein Barycenters
From MaRDI portal
Publication:150664
DOI10.48550/ARXIV.1310.4375arXiv1310.4375MaRDI QIDQ150664FDOQ150664
Authors: Marco Cuturi, Arnaud Doucet 
Publication date: 16 October 2013
Abstract: We present new algorithms to compute the mean of a set of empirical probability measures under the optimal transport metric. This mean, known as the Wasserstein barycenter, is the measure that minimizes the sum of its Wasserstein distances to each element in that set. We propose two original algorithms to compute Wasserstein barycenters that build upon the subgradient method. A direct implementation of these algorithms is, however, too costly because it would require the repeated resolution of large primal and dual optimal transport problems to compute subgradients. Extending the work of Cuturi (2013), we propose to smooth the Wasserstein distance used in the definition of Wasserstein barycenters with an entropic regularizer and recover in doing so a strictly convex objective whose gradients can be computed for a considerably cheaper computational cost using matrix scaling algorithms. We use these algorithms to visualize a large family of images and to solve a constrained clustering problem.
Cited In (1)
This page was built for publication: Fast Computation of Wasserstein Barycenters
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q150664)
