Informed Proposal Monte Carlo

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Informed Proposal Monte Carlo. / Khoshkholgh, Sarouyeh; Zunino, Andrea; Mosegaard, Klaus.

In: Geophysical Journal International, Vol. 226, No. 2, 01.08.2021, p. 1239-1248.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Khoshkholgh, S, Zunino, A & Mosegaard, K 2021, 'Informed Proposal Monte Carlo', Geophysical Journal International, vol. 226, no. 2, pp. 1239-1248. https://doi.org/10.1093/gji/ggab173

APA

Khoshkholgh, S., Zunino, A., & Mosegaard, K. (2021). Informed Proposal Monte Carlo. Geophysical Journal International, 226(2), 1239-1248. https://doi.org/10.1093/gji/ggab173

Vancouver

Khoshkholgh S, Zunino A, Mosegaard K. Informed Proposal Monte Carlo. Geophysical Journal International. 2021 Aug 1;226(2):1239-1248. https://doi.org/10.1093/gji/ggab173

Author

Khoshkholgh, Sarouyeh ; Zunino, Andrea ; Mosegaard, Klaus. / Informed Proposal Monte Carlo. In: Geophysical Journal International. 2021 ; Vol. 226, No. 2. pp. 1239-1248.

Bibtex

@article{31a8a01b757549c1830526b05a5fdea8,
title = "Informed Proposal Monte Carlo",
abstract = " Any search or sampling algorithm for solution of inverse problems needs guidance to be efficient. Many algorithms collect and apply information about the problem on the fly, and much improvement has been made in this way. However, as a consequence of the the No-Free-Lunch Theorem, the only way we can ensure a significantly better performance of search and sampling algorithms is to build in as much information about the problem as possible. In the special case of Markov Chain Monte Carlo sampling (MCMC) we review how this is done through the choice of proposal distribution, and we show how this way of adding more information about the problem can be made particularly efficient when based on an approximate physics model of the problem. A highly nonlinear inverse scattering problem with a high-dimensional model space serves as an illustration of the gain of efficiency through this approach. ",
keywords = "physics.geo-ph, stat.CO",
author = "Sarouyeh Khoshkholgh and Andrea Zunino and Klaus Mosegaard",
year = "2021",
month = aug,
day = "1",
doi = "10.1093/gji/ggab173",
language = "English",
volume = "226",
pages = "1239--1248",
journal = "Geophysical Journal International",
issn = "0956-540X",
publisher = "Oxford University Press",
number = "2",

}

RIS

TY - JOUR

T1 - Informed Proposal Monte Carlo

AU - Khoshkholgh, Sarouyeh

AU - Zunino, Andrea

AU - Mosegaard, Klaus

PY - 2021/8/1

Y1 - 2021/8/1

N2 - Any search or sampling algorithm for solution of inverse problems needs guidance to be efficient. Many algorithms collect and apply information about the problem on the fly, and much improvement has been made in this way. However, as a consequence of the the No-Free-Lunch Theorem, the only way we can ensure a significantly better performance of search and sampling algorithms is to build in as much information about the problem as possible. In the special case of Markov Chain Monte Carlo sampling (MCMC) we review how this is done through the choice of proposal distribution, and we show how this way of adding more information about the problem can be made particularly efficient when based on an approximate physics model of the problem. A highly nonlinear inverse scattering problem with a high-dimensional model space serves as an illustration of the gain of efficiency through this approach.

AB - Any search or sampling algorithm for solution of inverse problems needs guidance to be efficient. Many algorithms collect and apply information about the problem on the fly, and much improvement has been made in this way. However, as a consequence of the the No-Free-Lunch Theorem, the only way we can ensure a significantly better performance of search and sampling algorithms is to build in as much information about the problem as possible. In the special case of Markov Chain Monte Carlo sampling (MCMC) we review how this is done through the choice of proposal distribution, and we show how this way of adding more information about the problem can be made particularly efficient when based on an approximate physics model of the problem. A highly nonlinear inverse scattering problem with a high-dimensional model space serves as an illustration of the gain of efficiency through this approach.

KW - physics.geo-ph

KW - stat.CO

U2 - 10.1093/gji/ggab173

DO - 10.1093/gji/ggab173

M3 - Journal article

VL - 226

SP - 1239

EP - 1248

JO - Geophysical Journal International

JF - Geophysical Journal International

SN - 0956-540X

IS - 2

ER -

ID: 261065840