skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Accelerating Asymptotically Exact MCMC for Computationally Intensive Models via Local Approximations

Journal Article · · Journal of the American Statistical Association
 [1];  [1];  [2];  [3]
  1. Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States). Dept. of Aeronautics and Astronautics
  2. Harvard Univ., Cambridge, MA (United States). Dept. of Statistics
  3. Univ. of Ottawa, Ottawa (Canada). Dept. of Mathematics and Statistics
+ Show Author Affiliations

We construct a new framework for accelerating Markov chain Monte Carlo in posterior sampling problems where standard methods are limited by the computational cost of the likelihood, or of numerical models embedded therein. Our approach introduces local approximations of these models into the Metropolis–Hastings kernel, borrowing ideas from deterministic approximation theory, optimization, and experimental design. Previous efforts at integrating approximate models into inference typically sacrifice either the sampler’s exactness or efficiency; our work seeks to address these limitations by exploiting useful convergence characteristics of local approximations. We prove the ergodicity of our approximate Markov chain, showing that it samples asymptotically from the exact posterior distribution of interest. We describe variations of the algorithm that employ either local polynomial approximations or local Gaussian process regressors. Our theoretical results reinforce the key observation underlying this article: when the likelihood has some local regularity, the number of model evaluations per Markov chain Monte Carlo (MCMC) step can be greatly reduced without biasing the Monte Carlo average. Numerical experiments demonstrate multiple order-of-magnitude reductions in the number of forward model evaluations used in representative ordinary differential equation (ODE) and partial differential equation (PDE) inference problems, with both synthetic and real data. Supplementary materials for this article are available online.

Research Organization:
Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
Sponsoring Organization:
USDOE Office of Science (SC)
Grant/Contract Number:
SC0007099
OSTI ID:
1535376
Journal Information:
Journal of the American Statistical Association, Vol. 111, Issue 516; ISSN 0162-1459
Publisher:
Taylor & FrancisCopyright Statement
Country of Publication:
United States
Language:
English
Citation Metrics:
Cited by: 63 works
Citation information provided by
Web of Science

References (52)

Local Derivative-Free Approximation of Computationally Expensive Posterior Densities journal April 2012
Markov chain Monte Carlo Using an Approximation journal December 2005
Multivariate Adaptive Regression Splines journal March 1991
Bayesian Calibration and Uncertainty Analysis for Computationally Expensive Models Using Optimization and Radial Basis Function Approximation journal June 2008
Spline-Based Emulators for Radiative Shock Experiments With Measurement Error journal June 2013
Approximation of Bayesian Inverse Problems for PDEs journal January 2010
Bayesian Computation Using Design of Experiments-Based Interpolation Technique journal August 2012
Geometric convergence and central limit theorems for multidimensional Hastings and Metropolis algorithms journal March 1996
MCMC Methods for Functions: Modifying Old Algorithms to Make Them Faster journal August 2013
Smoothing by Local Regression: Principles and Methods book January 1996
Introduction to Derivative-Free Optimization book January 2009
Trust Region Methods book January 2000
High-Order Collocation Methods for Differential Equations with Random Inputs journal January 2005
Adaptive Construction of Surrogates for the Bayesian Solution of Inverse Problems journal January 2014
Adaptive Smolyak Pseudospectral Approximations journal January 2013
Approximate Bayesian computational methods journal October 2011
Stochastic spectral methods for efficient Bayesian solution of inverse problems journal June 2007
Uncertainty Quantification and Weak Approximation of an Elliptic Inverse Problem journal January 2011
Preconditioning Markov Chain Monte Carlo Simulations Using Coarse-Scale Models journal January 2006
Local Gaussian Process Approximation for Large Computer Experiments journal April 2015
The pseudo-marginal approach for efficient Monte Carlo computations journal April 2009
Sparse pseudospectral approximation method journal July 2012
A Stochastic Collocation Approach to Bayesian Inference in Inverse Problems journal January 2009
Robust Locally Weighted Regression and Smoothing Scatterplots journal December 1979
The Design and Analysis of Computer Experiments book January 2003
Bayesian calibration of computer models journal August 2001
Bayesian inverse problems with Monte Carlo forward models journal February 2013
A Sparse Grid Stochastic Collocation Method for Partial Differential Equations with Random Input Data journal January 2008
Approximating likelihoods for large spatial data sets journal May 2004
Efficient MCMC Schemes for Computationally Expensive Posterior Distributions journal February 2011
Convergence of adaptive and interacting Markov chain Monte Carlo algorithms journal December 2011
Construction of a genetic toggle switch in Escherichia coli journal January 2000
Design and Analysis of Computer Experiments journal November 1989
An Adaptive Metropolis Algorithm journal April 2001
Bayesian calibration of a large-scale geothermal reservoir model by a new adaptive delayed acceptance Metropolis Hastings algorithm: ADAPTIVE DELAYED ACCEPTANCE METROPOLIS-HASTINGS ALGORITHM journal October 2011
Parameter and State Model Reduction for Large-Scale Statistical Inverse Problems journal January 2010
General state space Markov chains and MCMC algorithms journal January 2004
Data Assimilation: A Mathematical Introduction book January 2015
Statistical inverse problems: Discretization, model reduction and inverse crimes journal January 2007
Design and analysis of computer experiments conference September 1998
The Design and Analysis of Computer Experiments book January 2018
Design and analysis of computer experiments journal December 2010
Local Gaussian Process Approximation for Large Computer Experiments text January 2015
Data Assimilation book January 2016
Local Gaussian process approximation for large computer experiments text January 2015
Stochastic spectral methods for efficient Bayesian solution of inverse problems
  • Marzouk, Youssef M.
  • BAYESIAN INFERENCE AND MAXIMUM ENTROPY METHODS IN SCIENCE AND ENGINEERING: 25th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering, AIP Conference Proceedings https://doi.org/10.1063/1.2149785
conference January 2005
The pseudo-marginal approach for efficient Monte Carlo computations text January 2009
MCMC Methods for Functions: Modifying Old Algorithms to Make Them Faster text January 2012
Adaptive Smolyak Pseudospectral Approximations preprint January 2012
Bayesian treed Gaussian process models with an application to computer modeling preprint January 2007
Uncertainty quantification and weak approximation of an elliptic inverse problem preprint January 2011
Austerity in MCMC Land: Cutting the Metropolis-Hastings Budget preprint January 2013

Cited By (8)

A transport-based multifidelity preconditioner for Markov chain Monte Carlo journal November 2019
Deterministic Sampling of Expensive Posteriors Using Minimum Energy Designs text January 2019
Deterministic Sampling of Expensive Posteriors Using Minimum Energy Designs text January 2019
Deterministic Sampling of Expensive Posteriors Using Minimum Energy Designs journal March 2019
Bayesian Estimation for Stochastic Gene Expression Using Multifidelity Models journal February 2019
Deterministic Sampling of Expensive Posteriors Using Minimum Energy Designs text January 2018
Surrogate-Based Bayesian Inverse Modeling of the Hydrological System: An Adaptive Approach Considering Surrogate Approximation Error text January 2018
A transport-based multifidelity preconditioner for Markov chain Monte Carlo preprint January 2018

Linked Research (3)

Similar Records

Itô-SDE MCMC method for Bayesian characterization of errors associated with data limitations in stochastic expansion methods for uncertainty quantification
Journal Article · Wed Nov 15 00:00:00 EST 2017 · Journal of Computational Physics · OSTI ID:1535376
Stochastic Engine Convergence Diagnostics
Technical Report · Tue Dec 11 00:00:00 EST 2001 · OSTI ID:1535376
Accelerating Markov Chain Monte Carlo sampling with diffusion models
Journal Article · Wed Dec 13 00:00:00 EST 2023 · Computer Physics Communications · OSTI ID:1535376
+3 more

Related Subjects

97 MATHEMATICS AND COMPUTING
Mathematics