Conditions for Rapid Mixing of Parallel and Simulated Tempering on Multimodal Distributions

dc.contributor.author

Woodard, DB

dc.contributor.author

Schmidler, SC

dc.contributor.author

Huber, M

dc.date.accessioned

2011-06-21T17:30:33Z

dc.date.available

2011-06-21T17:30:33Z

dc.date.issued

2009

dc.description.abstract

We give conditions under which a Markov chain constructed via parallel or simulated tempering is guaranteed to be rapidly mixing, which are applicable to a wide range of multimodal distributions arising in Bayesian statistical inference and statistical mechanics. We provide lower bounds on the spectral gaps of parallel and simulated tempering. These bounds imply a single set of sufficient conditions for rapid mixing of both techniques. A direct consequence of our results is rapid mixing of parallel and simulated tempering for several normal mixture models, and for the mean-field Ising model.

dc.description.version

Version of Record

dc.identifier.citation

Woodard,Dawn B.;Schmidler,Scott C.;Huber,Mark. 2009. Conditions for Rapid Mixing of Parallel and Simulated Tempering on Multimodal Distributions. Annals of Applied Probability 19(2): 617-640.

dc.identifier.issn

1050-5164

dc.identifier.uri

https://hdl.handle.net/10161/4407

dc.language.iso

en_US

dc.publisher

Institute of Mathematical Statistics

dc.relation.isversionof

10.1214/08-AAP555

dc.relation.journal

Annals of Applied Probability

dc.subject

markov chain monte carlo

dc.subject

tempering

dc.subject

rapidly mixing markov chains

dc.subject

spectral gap

dc.subject

metropolis algorithm

dc.subject

markov-chains

dc.subject

monte-carlo

dc.subject

convergence

dc.subject

statistics & probability

dc.title

Conditions for Rapid Mixing of Parallel and Simulated Tempering on Multimodal Distributions

dc.title.alternative
dc.type

Other article

duke.date.pubdate

2009-4-0

duke.description.issue

2

duke.description.volume

19

pubs.begin-page

617

pubs.end-page

640

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
283528700007.pdf
Size:
225.29 KB
Format:
Adobe Portable Document Format