Sökning: id:"swepub:oai:DiVA.org:liu-137624" >
Divide-and-Conquer ...
Divide-and-Conquer With Sequential Monte Carlo
-
- Lindsten, Fredrik, 1984- (författare)
- Uppsala universitet,Avdelningen för systemteknik,Reglerteknik
-
- Johansen, A. M. (författare)
- University of Warwick, England
-
- Andersson Naesseth, Christian, 1986- (författare)
- Linköpings universitet,Reglerteknik,Tekniska fakulteten
-
visa fler...
-
- Kirkpatrick, B. (författare)
- Intrepid Net Comp, MT USA
-
- Schön, Thomas B. (författare)
- Uppsala universitet,Avdelningen för systemteknik,Reglerteknik
-
- Aston, J. A. D. (författare)
- University of Cambridge, England
-
- Bouchard-Cote, A. (författare)
- University of British Columbia, Canada
-
visa färre...
-
(creator_code:org_t)
- 2017-04-24
- 2017
- Engelska.
-
Ingår i: Journal of Computational And Graphical Statistics. - : AMER STATISTICAL ASSOC. - 1061-8600 .- 1537-2715. ; 26:2, s. 445-458
- Relaterad länk:
-
http://arxiv.org/pdf...
-
visa fler...
-
https://urn.kb.se/re...
-
https://doi.org/10.1...
-
https://urn.kb.se/re...
-
visa färre...
Abstract
Ämnesord
Stäng
- We propose a novel class of Sequential Monte Carlo (SMC) algorithms, appropriate for inference in probabilistic graphical models. This class of algorithms adopts a divide-and-conquer approach based upon an auxiliary tree-structured decomposition of the model of interest, turning the overall inferential task into a collection of recursively solved subproblems. The proposed method is applicable to a broad class of probabilistic graphical models, including models with loops. Unlike a standard SMC sampler, the proposed divide-and-conquer SMC employs multiple independent populations of weighted particles, which are resampled, merged, and propagated as the method progresses. We illustrate empirically that this approach can outperform standard methods in terms of the accuracy of the posterior expectation and marginal likelihood approximations. Divide-and-conquer SMC also opens up novel parallel implementation options and the possibility of concentrating the computational effort on the most challenging subproblems. We demonstrate its performance on a Markov random field and on a hierarchical logistic regression problem. Supplementary materials including proofs and additional numerical results are available online.
Ämnesord
- NATURVETENSKAP -- Matematik -- Sannolikhetsteori och statistik (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Probability Theory and Statistics (hsv//eng)
Nyckelord
- Bayesian methods; Graphical models; Hierarchical models; Particle filters
Publikations- och innehållstyp
- ref (ämneskategori)
- art (ämneskategori)
Hitta via bibliotek
Till lärosätets databas