Tree-Based Diffusion Schrödinger Bridge with Applications to Wasserstein Barycenters

05/26/2023
by   Maxence Noble, et al.
0

Multi-marginal Optimal Transport (mOT), a generalization of OT, aims at minimizing the integral of a cost function with respect to a distribution with some prescribed marginals. In this paper, we consider an entropic version of mOT with a tree-structured quadratic cost, i.e., a function that can be written as a sum of pairwise cost functions between the nodes of a tree. To address this problem, we develop Tree-based Diffusion Schrödinger Bridge (TreeDSB), an extension of the Diffusion Schrödinger Bridge (DSB) algorithm. TreeDSB corresponds to a dynamic and continuous state-space counterpart of the multimarginal Sinkhorn algorithm. A notable use case of our methodology is to compute Wasserstein barycenters which can be recast as the solution of a mOT problem on a star-shaped tree. We demonstrate that our methodology can be applied in high-dimensional settings such as image interpolation and Bayesian fusion.

READ FULL TEXT

page 8

page 9

page 36

page 38

research
10/01/2021

On the complexity of the optimal transport problem with graph-structured cost

Multi-marginal optimal transport (MOT) is a generalization of optimal tr...
research
08/05/2022

Accelerating the Sinkhorn algorithm for sparse multi-marginal optimal transport by fast Fourier transforms

We consider the numerical solution of the discrete multi-marginal optima...
research
05/13/2022

Multi-Marginal Gromov-Wasserstein Transport and Barycenters

Gromov-Wasserstein (GW) distances are generalizations of Gromov-Haussdor...
research
03/31/2020

Inference with Aggregate Data: An Optimal Transport Approach

We consider inference problems over probabilistic graphical models with ...
research
09/08/2021

Fixed Support Tree-Sliced Wasserstein Barycenter

The Wasserstein barycenter has been widely studied in various fields, in...
research
11/30/2021

Path Integral Sampler: a stochastic control approach for sampling

We present Path Integral Sampler (PIS), a novel algorithm to draw sample...
research
06/16/2023

Building the Bridge of Schrödinger: A Continuous Entropic Optimal Transport Benchmark

Over the last several years, there has been a significant progress in de...

Please sign up or login with your details

Forgot password? Click here to reset