Computational Guarantees for Doubly Entropic Wasserstein Barycenters via Damped Sinkhorn Iterations
This work addresses computational challenges in optimal transport for researchers and practitioners, offering new guarantees for entropic barycenter algorithms, though it is incremental in building on existing regularization methods.
The authors tackled the problem of computing doubly regularized Wasserstein barycenters by proposing an algorithm based on damped Sinkhorn iterations with exact steps, which guarantees convergence for any regularization parameters. They also introduced an inexact variant with non-asymptotic convergence guarantees for approximating barycenters between discrete point clouds in a free-support setting.
We study the computation of doubly regularized Wasserstein barycenters, a recently introduced family of entropic barycenters governed by inner and outer regularization strengths. Previous research has demonstrated that various regularization parameter choices unify several notions of entropy-penalized barycenters while also revealing new ones, including a special case of debiased barycenters. In this paper, we propose and analyze an algorithm for computing doubly regularized Wasserstein barycenters. Our procedure builds on damped Sinkhorn iterations followed by exact maximization/minimization steps and guarantees convergence for any choice of regularization parameters. An inexact variant of our algorithm, implementable using approximate Monte Carlo sampling, offers the first non-asymptotic convergence guarantees for approximating Wasserstein barycenters between discrete point clouds in the free-support/grid-free setting.