DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Ensemble preconditioning for Markov chain Monte Carlo simulation

Abstract

We describe parallel Markov chain Monte Carlo methods that propagate a collective ensemble of paths, with local covariance information calculated from neighbouring replicas. Here, the use of collective dynamics eliminates multiplicative noise and stabilizes the dynamics, thus providing a practical approach to difficult anisotropic sampling problems in high dimensions. Numerical experiments with model problems demonstrate that dramatic potential speedups, compared to various alternative schemes, are attainable.

Authors:
; ORCiD logo;
Publication Date:
Research Org.:
Univ. of Chicago, IL (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
OSTI Identifier:
1344938
Alternate Identifier(s):
OSTI ID: 1502101
Grant/Contract Number:  
SC0014205
Resource Type:
Published Article
Journal Name:
Statistics and Computing
Additional Journal Information:
Journal Name: Statistics and Computing Journal Volume: 28 Journal Issue: 2; Journal ID: ISSN 0960-3174
Publisher:
Springer Science + Business Media
Country of Publication:
Germany
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Stochastic sampling; Markov chain Monte Carlo; MCMC; Computational statistics; Machine learning; BFGS; Langevin methods; Brownian dynamics

Citation Formats

Leimkuhler, Benedict, Matthews, Charles, and Weare, Jonathan. Ensemble preconditioning for Markov chain Monte Carlo simulation. Germany: N. p., 2017. Web. doi:10.1007/s11222-017-9730-1.
Leimkuhler, Benedict, Matthews, Charles, & Weare, Jonathan. Ensemble preconditioning for Markov chain Monte Carlo simulation. Germany. https://doi.org/10.1007/s11222-017-9730-1
Leimkuhler, Benedict, Matthews, Charles, and Weare, Jonathan. Mon . "Ensemble preconditioning for Markov chain Monte Carlo simulation". Germany. https://doi.org/10.1007/s11222-017-9730-1.
@article{osti_1344938,
title = {Ensemble preconditioning for Markov chain Monte Carlo simulation},
author = {Leimkuhler, Benedict and Matthews, Charles and Weare, Jonathan},
abstractNote = {We describe parallel Markov chain Monte Carlo methods that propagate a collective ensemble of paths, with local covariance information calculated from neighbouring replicas. Here, the use of collective dynamics eliminates multiplicative noise and stabilizes the dynamics, thus providing a practical approach to difficult anisotropic sampling problems in high dimensions. Numerical experiments with model problems demonstrate that dramatic potential speedups, compared to various alternative schemes, are attainable.},
doi = {10.1007/s11222-017-9730-1},
journal = {Statistics and Computing},
number = 2,
volume = 28,
place = {Germany},
year = {Mon Feb 27 00:00:00 EST 2017},
month = {Mon Feb 27 00:00:00 EST 2017}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
https://doi.org/10.1007/s11222-017-9730-1

Citation Metrics:
Cited by: 21 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Ensemble samplers with affine invariance
journal, January 2010

  • Goodman, Jonathan; Weare, Jonathan
  • Communications in Applied Mathematics and Computational Science, Vol. 5, Issue 1
  • DOI: 10.2140/camcos.2010.5.65

Population Monte Carlo
journal, December 2004

  • Cappé, O.; Guillin, A.; Marin, J. M.
  • Journal of Computational and Graphical Statistics, Vol. 13, Issue 4
  • DOI: 10.1198/106186004X12803

Free energy methods for Bayesian inference: efficient exploration of univariate Gaussian mixture posteriors
journal, June 2011

  • Chopin, Nicolas; Lelièvre, Tony; Stoltz, Gabriel
  • Statistics and Computing, Vol. 22, Issue 4
  • DOI: 10.1007/s11222-011-9257-9

Scaling limits for the transient phase of local Metropolis-Hastings algorithms
journal, April 2005

  • Christensen, Ole F.; Roberts, Gareth O.; Rosenthal, Jeffrey S.
  • Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 67, Issue 2
  • DOI: 10.1111/j.1467-9868.2005.00500.x

Hybrid Monte Carlo
journal, September 1987


Accelerating diffusions
journal, May 2005

  • Hwang, Chii-Ruey; Hwang-Ma, Shu-Yin; Sheu, Shuenn-Jyi
  • The Annals of Applied Probability, Vol. 15, Issue 2
  • DOI: 10.1214/105051605000000025

Philatelic Mixtures and Multimodal Densities
journal, December 1988


A general purpose sampling algorithm for continuous distributions (the t-walk)
journal, June 2010

  • Christen, J. Andrés; Fox, Colin
  • Bayesian Analysis, Vol. 5, Issue 2
  • DOI: 10.1214/10-BA603

On population-based simulation for static inference
journal, July 2007

  • Jasra, Ajay; Stephens, David A.; Holmes, Christopher C.
  • Statistics and Computing, Vol. 17, Issue 3
  • DOI: 10.1007/s11222-007-9028-9

A Stochastic Newton MCMC Method for Large-Scale Statistical Inverse Problems with Application to Seismic Inversion
journal, January 2012

  • Martin, James; Wilcox, Lucas C.; Burstedde, Carsten
  • SIAM Journal on Scientific Computing, Vol. 34, Issue 3
  • DOI: 10.1137/110845598

Population Monte Carlo algorithms.
journal, January 2001

  • Iba, Yukito
  • Transactions of the Japanese Society for Artificial Intelligence, Vol. 16
  • DOI: 10.1527/tjsai.16.279

Irreversible Langevin samplers and variance reduction: a large deviations approach
journal, May 2015


emcee : The MCMC Hammer
journal, March 2013

  • Foreman-Mackey, Daniel; Hogg, David W.; Lang, Dustin
  • Publications of the Astronomical Society of the Pacific, Vol. 125, Issue 925
  • DOI: 10.1086/670067

Multigrid Monte Carlo method. Conceptual foundations
journal, September 1989


Theoretical guarantees for approximate sampling from smooth and log-concave densities
journal, April 2016

  • Dalalyan, Arnak S.
  • Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 79, Issue 3
  • DOI: 10.1111/rssb.12183

Brownian dynamics as smart Monte Carlo simulation
journal, November 1978

  • Rossky, P. J.; Doll, J. D.; Friedman, H. L.
  • The Journal of Chemical Physics, Vol. 69, Issue 10
  • DOI: 10.1063/1.436415

Monte Carlo Calculation of the Average Extension of Molecular Chains
journal, February 1955

  • Rosenbluth, Marshall N.; Rosenbluth, Arianna W.
  • The Journal of Chemical Physics, Vol. 23, Issue 2
  • DOI: 10.1063/1.1741967

Metropolis Integration Schemes for Self-Adjoint Diffusions
journal, January 2014

  • Bou-Rabee, Nawaf; Donev, Aleksandar; Vanden-Eijnden, Eric
  • Multiscale Modeling & Simulation, Vol. 12, Issue 2
  • DOI: 10.1137/130937470

Stochastic Processes and Applications
book, January 2014


Variance Reduction Using Nonreversible Langevin Samplers
journal, March 2016

  • Duncan, A. B.; Lelièvre, T.; Pavliotis, G. A.
  • Journal of Statistical Physics, Vol. 163, Issue 3
  • DOI: 10.1007/s10955-016-1491-2

An Adaptive Metropolis Algorithm
journal, April 2001

  • Haario, Heikki; Saksman, Eero; Tamminen, Johanna
  • Bernoulli, Vol. 7, Issue 2
  • DOI: 10.2307/3318737

The computation of averages from equilibrium and nonequilibrium Langevin molecular dynamics
journal, January 2015

  • Leimkuhler, Benedict; Matthews, Charles; Stoltz, Gabriel
  • IMA Journal of Numerical Analysis
  • DOI: 10.1093/imanum/dru056

Riemann manifold Langevin and Hamiltonian Monte Carlo methods: Riemann Manifold Langevin and Hamiltonian Monte Carlo Methods
journal, March 2011

  • Girolami, Mark; Calderhead, Ben
  • Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 73, Issue 2
  • DOI: 10.1111/j.1467-9868.2010.00765.x

Improved Diffusion Monte Carlo
journal, June 2014

  • Hairer, Martin; Weare, Jonathan
  • Communications on Pure and Applied Mathematics, Vol. 67, Issue 12
  • DOI: 10.1002/cpa.21526

Faster estimation of Bayesian models in ecology using Hamiltonian Monte Carlo
journal, November 2016

  • Monnahan, Cole C.; Thorson, James T.; Branch, Trevor A.
  • Methods in Ecology and Evolution, Vol. 8, Issue 3
  • DOI: 10.1111/2041-210X.12681

Works referencing / citing this record:

A new bi‐fidelity model reduction method for Bayesian inverse problems
journal, May 2019

  • Ou, Na; Jiang, Lijian; Lin, Guang
  • International Journal for Numerical Methods in Engineering, Vol. 119, Issue 10
  • DOI: 10.1002/nme.6079