The Library
Hybrid Monte Carlo on Hilbert spaces
Tools
Beskos, Alexandros, Pinski, F. J., SanzSerna, J. M. and Stuart, A. M.. (2011) Hybrid Monte Carlo on Hilbert spaces. Stochastic Processes and their Applications, Vol.121 (No.10). pp. 22012230. ISSN 03044149

PDF
WRAP_Stuart_Hybrid.pdf  Published Version  Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader Download (624Kb) 
Official URL: http://dx.doi.org/10.1016/j.spa.2011.06.003
Abstract
The Hybrid Monte Carlo (HMC) algorithm provides a framework for sampling from complex, highdimensional target distributions. In contrast with standard Markov chain Monte Carlo (MCMC) algorithms, it generates nonlocal, nonsymmetric moves in the state space, alleviating random walk type behaviour for the simulated trajectories. However, similarly to algorithms based on random walk or Langevin proposals, the number of steps required to explore the target distribution typically grows with the dimension of the state space. We define a generalized HMC algorithm which overcomes this problem for target measures arising as finitedimensional approximations of measures pi which have density with respect to a Gaussian measure on an infinitedimensional Hilbert space. The key idea is to construct an MCMC method which is well defined on the Hilbert space itself. We successively address the following issues in the infinitedimensional setting of a Hilbert space: (i) construction of a probability measure Pi in an enlarged phase space having the target pi as a marginal, together with a Hamiltonian flow that preserves Pi; (ii) development of a suitable geometric numerical integrator for the Hamiltonian flow; and (iii) derivation of an accept/reject rule to ensure preservation of Pi when using the above numerical integrator instead of the actual Hamiltonian flow. Experiments are reported that compare the new algorithm with standard HMC and with a version of the Langevin MCMC method defined on a Hilbert space.
[error in script] [error in script]Item Type:  Journal Article 

Subjects:  Q Science > QA Mathematics 
Divisions:  Faculty of Science > Mathematics 
Library of Congress Subject Headings (LCSH):  Monte Carlo method, Hilbert space 
Journal or Publication Title:  Stochastic Processes and their Applications 
Publisher:  Elsevier BV * NorthHolland 
ISSN:  03044149 
Date:  2011 
Volume:  Vol.121 
Number:  No.10 
Page Range:  pp. 22012230 
Identification Number:  10.1016/j.spa.2011.06.003 
Status:  Peer Reviewed 
Publication Status:  Published 
Access rights to Published version:  Restricted or Subscription Access 
Funder:  Spain. Ministerio de Ciencia e Innovación (MICINN), Engineering and Physical Sciences Research Council (EPSRC), European Research Council (ERC) 
Grant number:  MTM201018246C0301 (MICINN) 
References:  [1] A. Beskos, N.S. Pillai, G.O. Roberts, J.M. SanzSerna, A.M. Stuart, optimal tuning of hybrid MonteCarlo, Technical Report (2010) (submitted for publication). [2] A. Beskos, G.O. Roberts, A.M. Stuart, Optimal scalings for local Metropolis–Hastings chains on nonproduct targets in high dimensions, Ann. Appl. Probab. 19 (2009) 863–898. [3] A. Beskos, G.O. Roberts, A.M. Stuart, J. Voss, MCMC methods for diffusion bridges, Stoch. Dyn. 8 (2008) 319–350. [4] A. Beskos, A.M. Stuart, MCMC methods for sampling function space, in: R. Jeltsch, G. Wanner (Eds.), Invited Lectures, Sixth International Congress on Industrial and Applied Mathematics, ICIAM07, European Mathematical Society, 2009, pp. 337–364. [5] A. Beskos, A.M. Stuart, Computational complexity of Metropolis–Hastings methods in high dimensions, in: P. L’Ecuyer, A.B. Owen (Eds.), Monte Carlo and QuasiMonte Carlo Methods 2008, SpringerVerlag, 2010. [6] V. Bogachev, Gaussian Measures, in: Mathematical Surveys and Monographs, vol. 62, American Mathematical Society, 1998. [7] J. Bourgain, Periodic nonlinear Schr¨odinger equation and invariant measures, Comm. Math. Phys. 166 (1994) 1–26. [8] G. Da Prato, J. Zabczyk, Stochastic Equations in Infinite Dimensions, in: Encyclopedia of Mathematics and its Applications, vol. 44, Cambridge University Press, Cambridge, 1992. [9] S. Duane, A.D. Kennedy, B. Pendleton, D. Roweth, Hybrid Monte Carlo, Phys. Lett. B 195 (1987) 216–222. [10] M. Girolami, B. Calderhead, Riemann manifold Langevin and Hamiltonian Monte Carlo methods, J. R. Stat. Soc. Ser. B Stat. Methodol. 73 (2011) 123–214. [11] E. Hairer, C. Lubich, G. Wanner, Geometric Numerical Integration. StructurePreserving Algorithms for Ordinary Differential Equations, second ed., in: Springer Series in Computational Mathematics, vol. 31, SpringerVerlag, Berlin, 2006. [12] M. Hairer, A.M. Stuart, J. Voss, Analysis of spdes arising in path sampling part II: the nonlinear case, Ann. Appl. Probab. 17 (2010) 1657–1706. [13] M. Hairer, A.M. Stuart, J. Voss, Signal processing problems on function space: Bayesian formulation, stochastic PDEs and effective MCMC methods, in: D. Crisan, B. Rozovsky (Eds.), Oxford Handbook of Nonlinear Filtering, 2011. [14] A. Irback, Hybrid Monte Carlo simulation of polymer chains, J. Chem. Phys. 101 (1994) 1661–1667. [15] J.L. Lebowitz, H.A. Rose, E.R. Speer, Statistical mechanics of the nonlinear Schr¨odinger equation, J. Stat. Phys. 50 (1988) 657–687. [16] M.A. Lifshits, Gaussian Random Functions, in: Mathematics and its Applications, vol. 322, Kluwer, 1995. [17] J.S. Liu, Monte Carlo Strategies in Scientific Computing, in: Springer Series in Statistics, Springer, New York, 2008. [18] B. Mehlig, D.W. Heermann, B.M. Forrest, Exaxct Langevin algorithms, Mol. Phys. 8 (1992) 1347–1357. [19] R.M. Neal, Probabilistic inference using markov chain Monte Carlo methods, Technical Report, Department of Computer Science, University of Toronto, 1993. [20] R.M. Neal, MCMC using Hamiltonian dynamics, in: S. Brooks, A. Gelman, G. Jones, X.L. Meng (Eds.), Handbook of Markov Chain Monte Carlo, 2011. [21] K.R. Parthasarathy, Probability Measures on Metric Spaces, in: Probability and Mathematical Statistics, vol. 3, Academic Press Inc., New York, 1967. [22] G.O. Roberts, A. Gelman, W.R. Gilks, Weak convergence and optimal scaling of random walk Metropolis algorithms, Ann. Appl. Probab. 7 (1997) 110–120. [23] G.O. Roberts, J.S. Rosenthal, Optimal scaling of discrete approximations to Langevin diffusions, J. R. Stat. Soc. Ser. B Stat. Methodol. 60 (1998) 255–268. [24] J.C. Robinson, Infinite Dimensional Dynamical Systems, in: Cambridge Texts in Applied Mathematics, Cambridge University Press, 2001. [25] J.M. SanzSerna, M.P. Calvo, Numerical Hamiltonian Problems, in: Applied Mathematics and Mathematical Computation, vol. 7, Chapman & Hall, London, 1994. [26] C. Sch¨utte, Conformational dynamics: modelling, theory, algorithm, and application to biomolecules (1998). Habilitation Thesis, Dept. of Mathematics and Computer Science, Free University Berlin. Available at: http://proteomicsberlin.de/89/. [27] A.M. Stuart, Inverse problems: a Bayesian approach, Acta Numer. 19 (2010). 
URI:  http://wrap.warwick.ac.uk/id/eprint/38357 
Data sourced from Thomson Reuters' Web of Knowledge
Actions (login required)
View Item 