Entropic optimal transport: convergence of potentials

DOI10.1007/S00440-021-01096-8MaRDI QIDQ2089759zbMATH OpenOpenAlexFDO

Authors Marcel Nutz, Johannes Wiesel

Publication date 24 October 2022

Published in Zeitschrift für Wahrscheinlichkeitstheorie und Verwandte Gebiete (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2104.11720

zbMATH Keywords

optimal transportentropic regularizationSchrödinger potentials

Mathematics Subject Classification ID

Convex programming (90C25) Special problems of linear programming (transportation, multi-index, data envelopment analysis, etc.) (90C08) Linear optimal control problems (49N05)

Abstract: We study the potential functions that determine the optimal density for $varepsilon$-entropically regularized optimal transport, the so-called Schr"odinger potentials, and their convergence to the counterparts in classical optimal transport, the Kantorovich potentials. In the limit $varepsilono0$ of vanishing regularization, strong compactness holds in $L^1$ and cluster points are Kantorovich potentials. In particular, the Schr"odinger potentials converge in $L^1$ to the Kantorovich potentials as soon as the latter are unique. These results are proved for all continuous, integrable cost functions on Polish spaces. In the language of Schr"odinger bridges, the limit corresponds to the small-noise regime.

Recommendations

• Entropic optimal transport: geometry and large deviations
• Stability of Schrödinger potentials and convergence of Sinkhorn's algorithm
• Convergence rate of general entropic optimal transport costs
• Entropic regularization of continuous optimal transport problems
• An optimal transport approach for the Schrödinger bridge problem and convergence of Sinkhorn algorithm

Cites work

• scientific article; zbMATH DE number 4080537 (Why is no real title available?)
• A formula for the time derivative of the entropic cost and applications
• A survey of the Schrödinger problem and some of its connections with optimal transport
• An automorphism of product measures
• An optimal transport approach for the Schrödinger bridge problem and convergence of Sinkhorn algorithm
• Asymptotic analysis of the exponential penalty trajectory in linear programming
• Barycenters in the Wasserstein space
• Closedness of sum spaces and the generalized Schrödinger problem
• Computational optimal transport. With applications to data sciences
• Convergence of entropic schemes for optimal transport and gradient flows
• Convex duality in nonlinear optimal transport
• Decomposition of Multivariate Functions
• Entropy minimization and Schrödinger processes in infinite dimensions
• Entropy minimization, \(DAD\) problems, and doubly stochastic kernels
• From the Schrödinger problem to the Monge-Kantorovich problem
• I-divergence geometry of probability distributions and minimization problems
• Large deviations techniques and applications.
• Note on the Schrödinger equation and \(I\)-projections
• Optimal Transport
• Second order differentiation formula on \(\mathsf{RCD}^*(K,N)\) spaces
• The Sinkhorn algorithm, parabolic optimal transport and geometric Monge-Ampère equations

Cited in

(28)

• Gromov-Wasserstein distances: entropic regularization, duality and sample complexity
• Stability of Schrödinger potentials and convergence of Sinkhorn's algorithm
• Gradient estimates for the Schrödinger potentials: convergence to the Brenier map and quantitative stability
• An external dual charge approach to the multimarginal optimal transport with Coulomb cost
• Entropic optimal transport: geometry and large deviations
• Long-time behaviour of entropic interpolations
• Domain decomposition for entropy regularized optimal transport
• Limit theorems for entropic optimal transport maps and Sinkhorn divergence
• On entropy martingale optimal transport theory
• Weak semiconvexity estimates for Schrödinger potentials and logarithmic Sobolev inequality for Schrödinger bridges
• Statistical inference with regularized optimal transport
• Convergence of entropic schemes for optimal transport and gradient flows
• On the Degeneracy of Optimal Transportation
• Quantitative stability of regularized optimal transport and convergence of Sinkhorn's algorithm
• Entropy martingale optimal transport and nonlinear pricing-hedging duality
• Martingale Schrödinger bridges and optimal semistatic portfolios
• scientific article; zbMATH DE number 5292626 (Why is no real title available?)
• An Improved Central Limit Theorem and Fast Convergence Rates for Entropic Transportation Costs
• A strong duality principle for equivalence couplings and total variation
• Propagation of chaos for mean field Schrödinger problems
• Regularizing effects of the entropy functional in optimal transport and planning problems
• Entropic Optimal Planning for Path-Dependent Mean Field Games
• Limit distributions and sensitivity analysis for empirical entropic optimal transport on countable spaces
• On the difference between entropic cost and the optimal transport cost
• Stability and sample complexity of divergence regularized optimal transport
• Convergence rate of general entropic optimal transport costs
• Stability of entropic optimal transport and Schrödinger bridges
• About the analogy between optimal transport and minimal entropy

This page was built for publication: Entropic optimal transport: convergence of potentials

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2089759)