Accelerating Asymptotically Exact MCMC for Computationally Intensive Models via Local Approximations
Name
1402.1694v4.pdf
Size
2.61 MB
Format
Adobe PDF
Checksum (MD5)
0f8fb8d06e41c345d14b88e539e22ee1
Author(s)
Conrad, Patrick R.
Marzouk, Youssef M.
Pillai, Natesh S.
Smith, Aaron
Date Issued
October 2015
Journal
Journal of the American Statistical Association
Publisher
American Statistical Association
Citation
Conrad, Patrick R., Youssef M. Marzouk, Natesh S. Pillai, and Aaron Smith. “Accelerating Asymptotically Exact MCMC for Computationally Intensive Models via Local Approximations.” Journal of the American Statistical Association (October 21, 2015): 00–00.
Version
Author's final manuscript
Abstract
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 paper: when the likelihood has some local regularity, the number of model evaluations per 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 ODE and PDE inference problems, with both synthetic and real data.
MIT Department
Massachusetts Institute of Technology. Department of Aeronautics and Astronautics
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1080/01621459.2015.1096787
