Diffeomorphic Random Sampling Using Optimal Information Transport

In this article we explore an algorithm for diffeomorphic random sampling of nonuniform probability distributions on Riemannian manifolds. The algorithm is based on optimal information transport (OIT)—an analogue of optimal mass transport (OMT). Our framework uses the deep geometric connections between the Fisher-Rao metric on the space of probability densities and the right-invariant information metric on the group of diffeomorphisms. The resulting sampling algorithm is a promising alternative to OMT, in particular as our formulation is semi-explicit, free of the nonlinear Monge–Ampere equation. Compared to Markov Chain Monte Carlo methods, we expect our algorithm to stand up well when a large number of samples from a low dimensional nonuniform distribution is needed.

[1]  Shun-ichi Amari,et al.  Methods of information geometry , 2000 .

[2]  L. Younes,et al.  On the metrics and euler-lagrange equations of computational anatomy. , 2002, Annual review of biomedical engineering.

[3]  Martin Bauer,et al.  Uniqueness of the Fisher–Rao metric on the space of smooth densities , 2014, 1411.5577.

[4]  Richard S. Hamilton,et al.  The inverse function theorem of Nash and Moser , 1982 .

[5]  B. Khesin,et al.  Geometry of Diffeomorphism Groups, Complete integrability and Geometric statistics , 2013 .

[6]  Martin Bauer,et al.  Diffeomorphic Density Matching by Optimal Information Transport , 2015, SIAM J. Imaging Sci..

[7]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[8]  Thomas Friedrich,et al.  Die Fisher‐Information und symplektische Strukturen , 1991 .

[9]  Klas Modin,et al.  Generalized Hunter–Saxton Equations, Optimal Information Transport, and Factorization of Diffeomorphisms , 2012, 1203.4463.

[10]  J. Moser.  On the volume elements on a manifold , 1965 .

[11]  Youssef M. Marzouk,et al.  Bayesian inference with optimal maps , 2011, J. Comput. Phys..

[12]  Sebastian Reich,et al.  A Nonparametric Ensemble Transform Method for Bayesian Inference , 2012, SIAM J. Sci. Comput..

[13]  Y. Marzouk,et al.  An introduction to sampling via measure transport , 2016, 1602.05023.

[14]  F. Otto.  THE GEOMETRY OF DISSIPATIVE EVOLUTION EQUATIONS: THE POROUS MEDIUM EQUATION , 2001 .

[15]  C. Villani.  Optimal Transport: Old and New , 2008 .

[16]  B. Khesin,et al.  The Geometry of Infinite-Dimensional Groups , 2009 .