Small-Noise Analysis and Symmetrization of Implicit Monte Carlo Samplers

Jonathan Goodman, Kevin K. Lin, Matthias Morzfeld

Research output: Contribution to journalArticle

Abstract

Implicit samplers are algorithms for producing independent, weighted samples from multivariate probability distributions. These are often applied in Bayesian data assimilation algorithms. We use Laplace asymptotic expansions to analyze two implicit samplers in the small noise regime. Our analysis suggests a symmetrization of the algorithms that leads to improved implicit sampling schemes at a relatively small additional cost. Computational experiments confirm the theory and show that symmetrization is effective for small noise sampling problems.

Original languageEnglish (US)
JournalCommunications on Pure and Applied Mathematics
DOIs
StateAccepted/In press - 2015

Fingerprint

Symmetrization
Sampling
Data Assimilation
Multivariate Distribution
Laplace
Computational Experiments
Probability distributions
Asymptotic Expansion
Probability Distribution
Costs
Experiments

ASJC Scopus subject areas

  • Mathematics(all)
  • Applied Mathematics

Cite this

Small-Noise Analysis and Symmetrization of Implicit Monte Carlo Samplers. / Goodman, Jonathan; Lin, Kevin K.; Morzfeld, Matthias.

In: Communications on Pure and Applied Mathematics, 2015.

Research output: Contribution to journalArticle

@article{3c5d0bde9ee9431a96b04e967c8bf020,
title = "Small-Noise Analysis and Symmetrization of Implicit Monte Carlo Samplers",
abstract = "Implicit samplers are algorithms for producing independent, weighted samples from multivariate probability distributions. These are often applied in Bayesian data assimilation algorithms. We use Laplace asymptotic expansions to analyze two implicit samplers in the small noise regime. Our analysis suggests a symmetrization of the algorithms that leads to improved implicit sampling schemes at a relatively small additional cost. Computational experiments confirm the theory and show that symmetrization is effective for small noise sampling problems.",
author = "Jonathan Goodman and Lin, {Kevin K.} and Matthias Morzfeld",
year = "2015",
doi = "10.1002/cpa.21592",
language = "English (US)",
journal = "Communications on Pure and Applied Mathematics",
issn = "0010-3640",
publisher = "Wiley-Liss Inc.",

}

TY - JOUR

T1 - Small-Noise Analysis and Symmetrization of Implicit Monte Carlo Samplers

AU - Goodman, Jonathan

AU - Lin, Kevin K.

AU - Morzfeld, Matthias

PY - 2015

Y1 - 2015

N2 - Implicit samplers are algorithms for producing independent, weighted samples from multivariate probability distributions. These are often applied in Bayesian data assimilation algorithms. We use Laplace asymptotic expansions to analyze two implicit samplers in the small noise regime. Our analysis suggests a symmetrization of the algorithms that leads to improved implicit sampling schemes at a relatively small additional cost. Computational experiments confirm the theory and show that symmetrization is effective for small noise sampling problems.

AB - Implicit samplers are algorithms for producing independent, weighted samples from multivariate probability distributions. These are often applied in Bayesian data assimilation algorithms. We use Laplace asymptotic expansions to analyze two implicit samplers in the small noise regime. Our analysis suggests a symmetrization of the algorithms that leads to improved implicit sampling schemes at a relatively small additional cost. Computational experiments confirm the theory and show that symmetrization is effective for small noise sampling problems.

UR - http://www.scopus.com/inward/record.url?scp=84935077144&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84935077144&partnerID=8YFLogxK

U2 - 10.1002/cpa.21592

DO - 10.1002/cpa.21592

M3 - Article

JO - Communications on Pure and Applied Mathematics

JF - Communications on Pure and Applied Mathematics

SN - 0010-3640

ER -