Conference item
Interacting particle Markov chain Monte Carlo
- Abstract:
- We introduce interacting particle Markov chain Monte Carl (iPMCMC), a PMCMC method that introduces a coupling between multiple standard and conditional sequential Monte Carlo samplers. Like related methods, iPMCMC is a Markov chain Monte Carlo sampler on an extended space. We present empirical results that show significant improvements in mixing rates relative to both non-interacting PMCMC samplers and a single PMCMC sampler with an equivalent total computational budget. An additional advantage of the iPMCMC method is that it is suitable for distributed and multi-core architectures.
- Publication status:
- Published
Actions
Authors
+ US Air Force Research Laboratory
More from this funder
- Funding agency for:
- Wood, F
- Grant:
- DARPA PPAML Cooperative Agreement number FA8750-14-2-0006, Sub Award number 61160290-111668
- Publisher:
- Journal of Machine Learning Research
- Host title:
- ICML 2016: 33rd International Conference on Machine Learning
- Journal:
- ICML 2016: 33rd International Conference on Machine Learning More from this journal
- Publication date:
- 2016-06-11
- Acceptance date:
- 2016-04-24
- ISSN:
-
1533-7928
- Pubs id:
-
pubs:624354
- UUID:
-
uuid:cf43029b-7133-402a-83f7-314f10aa91d8
- Local pid:
-
pubs:624354
- Source identifiers:
-
624354
- Deposit date:
-
2016-05-27
Terms of use
- Copyright holder:
- Rainforth et al
- Copyright date:
- 2016
- Notes:
- JMLR: WandCP volume 48. Copyright 2016 by the author(s). This article was presented at ICML 2016: 33rd International Conference on Machine Learning. Available online at [http://www.jmlr.org/proceedings/papers/v48/rainforth16.html]
If you are the owner of this record, you can report an update to it here: Report update to this record