Geometric MCMC for infinite-dimensional inverse problems

JOURNAL OF COMPUTATIONAL PHYSICS

卷:335

Article

Beskos, Alexandros¹  Girolami, Mark^2,3  Lan, Shiwei⁴  Farrell, Patrick E.^5,6  Stuart, Andrew M.⁴ 

[1] UCL, Dept Stat Sci, Gower St, London WC1E 6BT, England

[2] Imperial Coll London, Dept Math, London SW7 2AZ, England

[3] Alan Turing Inst Data Sci, British Lib, 96 Euston Rd, London NW1 2DB, England

[4] CALTECH, Dept Comp Math Sci, Pasadena, CA 91125 USA

[5] Univ Oxford, Math Inst, Radcliffe Observ Quarter, Andrew Wiles Bldg,Woodstock Rd, Oxford OX2 6GG, England

[6] Simula Res Lab, Ctr Biomed Comp, Martin Linges Vei 17, N-1364 Fornebu, Norway

关键词: Markov chain Monte Carlo;    Local preconditioning;    Infinite dimensions;    Bayesian inverse problems;    Uncertainty quantification;   

DOI  :  10.1016/j.jcp.2016.12.041

来源: Elsevier

【 摘 要 】

Bayesian inverse problems often involve sampling posterior distributions on infinite dimensional function spaces. Traditional Markov chain Monte Carlo (MCMC) algorithms are characterized by deteriorating mixing times upon mesh-refinement, when the finite dimensional approximations become more accurate. Such methods are typically forced to reduce step-sizes as the discretization gets finer, and thus are expensive as a function of dimension. Recently, a new class of MCMC methods with mesh-independent convergence times has emerged. However, few of them take into account the geometry of the posterior informed by the data. At the same time, recently developed geometric MCMC algorithms have been found to be powerful in exploring complicated distributions that deviate significantly from elliptic Gaussian laws, but are in general computationally intractable for models defined in infinite dimensions. In this work, we combine geometric methods on a finite-dimensional subspace with mesh-independent infinite-dimensional approaches. Our objective is to speed up MCMC mixing times, without significantly increasing the computational cost per step (for instance, in comparison with the vanilla preconditioned Crank Nicolson (pCN) method). This is achieved by using ideas from geometric MCMC to probe the complex structure of an intrinsic finite-dimensional subspace where most data information concentrates, while retaining robust mixing times as the dimension grows by using pCN-like methods in the complementary subspace. The resulting algorithms are demonstrated in the context of three challenging inverse problems arising in subsurface flow, heat conduction and incompressible flow control. The algorithms exhibit up to two orders of magnitude improvement in sampling efficiency when compared with the pCN method. (C) 2017 The Authors. Published by Elsevier Inc.

【 授权许可】

Free