vix.ing · top · new · best · stats

Multi-marginal Schrodinger bridges

2019/02/22 by Yongxin Chen, Chen, Yongxin, Giovanni Conforti +5 · 3 citations
Computer Science · Engineering · Mathematics · #46Nxx #58J65 #93E20 #Anomaly Detection Techniques and Applications #Applied mathematics #Computer science #Dimension (graph theory) #FOS: Mathematics #Flow (mathematics) #Fluid Dynamics and Turbulent Flows #Geometry #Image (mathematics) #Interpolation (computer graphics) #Joint probability distribution #Marginal distribution #Mathematical analysis #Mathematics #Measure (data warehouse) #Optimization and Control (math.OC) #Probability (math.PR) #Probability measure #Pure mathematics #Random variable #Statistical Methods and Inference #math.OC #math.PR #msc:46Nxx #msc:58J65 #msc:93E20

paper · pdf · doi:10.48550/arxiv.1902.08319

published in arXiv (Cornell University) (Cornell University) · 8 pages

arxiv created 2019/02/22 · openalex publication_date 2019/02/22 · arxiv updated 2019/02/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/04

Abstract

We consider the problem to identify the most likely flow in phase space, of (inertial) particles under stochastic forcing, that is in agreement with spatial (marginal) distributions that are specified at a set of points in time. The question raised generalizes the classical Schrodinger Bridge Problem (SBP) which seeks to interpolate two specified end-point marginal distributions of overdamped particles driven by stochastic excitation. While we restrict our analysis to second-order dynamics for the particles, the data represents partial (i.e., only positional) information on the flow at \em multiple time-points. The solution sought, as in SBP, represents a probability law on the space of paths this closest to a uniform prior while consistent with the given marginals. We approach this problem as an optimal control problem to minimize an action integral a la Benamou-Brenier, and derive a time-symmetric formulation that includes a Fisher information term on the velocity field. We underscore the relation of our problem to recent measure-valued splines in Wasserstein space, which is akin to that between SBP and Optimal Mass Transport (OMT). The connection between the two provides a Sinkhorn-like approach to computing measure-valued splines. We envision that interpolation between measures as sought herein will have a wide range of applications in signal/images processing as well as in data science in cases where data have a temporal dimension.

Cited by

Related