A dynamic programming approach to efficient sampling from Boltzmann distributions

Archis Ghate; Robert L. Smith

doi:10.1016/j.orl.2008.07.009

A dynamic programming approach to efficient sampling from Boltzmann distributions

Archis Ghate, Robert L. Smith

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

Markov chain methods for Boltzmann sampling work in phases with decreasing temperatures. The number of transitions in each phase crucially affects terminal state distribution. We employ dynamic programming to allocate iterations to phases to improve guarantees on sample quality. Numerical experiments on the Ising model are presented.

Original language	English (US)
Pages (from-to)	665-668
Number of pages	4
Journal	Operations Research Letters
Volume	36
Issue number	6
DOIs	https://doi.org/10.1016/j.orl.2008.07.009
State	Published - Nov 2008
Externally published	Yes

Keywords

Cooling schedule
Markov chain Monte Carlo
Simulated Annealing
Warm start

Publisher link

10.1016/j.orl.2008.07.009

Find It

Find It at U of M Libraries

Cite this

@article{7187b4889754478c9d3256a70570dfbd,

title = "A dynamic programming approach to efficient sampling from Boltzmann distributions",

abstract = "Markov chain methods for Boltzmann sampling work in phases with decreasing temperatures. The number of transitions in each phase crucially affects terminal state distribution. We employ dynamic programming to allocate iterations to phases to improve guarantees on sample quality. Numerical experiments on the Ising model are presented.",

keywords = "Cooling schedule, Markov chain Monte Carlo, Simulated Annealing, Warm start",

author = "Archis Ghate and Smith, {Robert L.}",

year = "2008",

month = nov,

doi = "10.1016/j.orl.2008.07.009",

language = "English (US)",

volume = "36",

pages = "665--668",

journal = "Operations Research Letters",

issn = "0167-6377",

publisher = "Elsevier",

number = "6",

}

TY - JOUR

T1 - A dynamic programming approach to efficient sampling from Boltzmann distributions

AU - Ghate, Archis

AU - Smith, Robert L.

PY - 2008/11

Y1 - 2008/11

N2 - Markov chain methods for Boltzmann sampling work in phases with decreasing temperatures. The number of transitions in each phase crucially affects terminal state distribution. We employ dynamic programming to allocate iterations to phases to improve guarantees on sample quality. Numerical experiments on the Ising model are presented.

AB - Markov chain methods for Boltzmann sampling work in phases with decreasing temperatures. The number of transitions in each phase crucially affects terminal state distribution. We employ dynamic programming to allocate iterations to phases to improve guarantees on sample quality. Numerical experiments on the Ising model are presented.

KW - Cooling schedule

KW - Markov chain Monte Carlo

KW - Simulated Annealing

KW - Warm start

UR - http://www.scopus.com/inward/record.url?scp=54249157349&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=54249157349&partnerID=8YFLogxK

U2 - 10.1016/j.orl.2008.07.009

DO - 10.1016/j.orl.2008.07.009

M3 - Article

AN - SCOPUS:54249157349

SN - 0167-6377

VL - 36

SP - 665

EP - 668

JO - Operations Research Letters

JF - Operations Research Letters

IS - 6

ER -

A dynamic programming approach to efficient sampling from Boltzmann distributions

Abstract

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this