Particle Gibbs split-merge sampling for Bayesian inference in mixture models
This paper presents an original Markov chain Monte Carlo method to sample from the posterior distribution of conjugate mixture models. This algorithm relies on a flexible split-merge procedure built using the particle Gibbs sampler introduced in Andrieu et al. (2009, 2010). The resulting so-called P...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Published: |
Journal of Machine Learning Research
2017
|
| _version_ | 1797072561036591104 |
|---|---|
| author | Bouchard-Côté, A Doucet, A Roth, A |
| author_facet | Bouchard-Côté, A Doucet, A Roth, A |
| author_sort | Bouchard-Côté, A |
| collection | OXFORD |
| description | This paper presents an original Markov chain Monte Carlo method to sample from the posterior distribution of conjugate mixture models. This algorithm relies on a flexible split-merge procedure built using the particle Gibbs sampler introduced in Andrieu et al. (2009, 2010). The resulting so-called Particle Gibbs Split-Merge sampler does not require the computation of a complex acceptance ratio and can be implemented using existing sequential Monte Carlo libraries. We investigate its performance experimentally on synthetic problems as well as on geolocation data. Our results show that for a given computational budget, the Particle Gibbs Split-Merge sampler empirically outperforms existing split merge methods. The code and instructions allowing to reproduce the experiments is available at https://github.com/aroth85/pgsm. |
| first_indexed | 2024-03-06T23:09:29Z |
| format | Journal article |
| id | oxford-uuid:64f62e09-fe01-4328-833a-a372e2ca2268 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T23:09:29Z |
| publishDate | 2017 |
| publisher | Journal of Machine Learning Research |
| record_format | dspace |
| spelling | oxford-uuid:64f62e09-fe01-4328-833a-a372e2ca22682022-03-26T18:22:27ZParticle Gibbs split-merge sampling for Bayesian inference in mixture modelsJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:64f62e09-fe01-4328-833a-a372e2ca2268Symplectic Elements at OxfordJournal of Machine Learning Research2017Bouchard-Côté, ADoucet, ARoth, AThis paper presents an original Markov chain Monte Carlo method to sample from the posterior distribution of conjugate mixture models. This algorithm relies on a flexible split-merge procedure built using the particle Gibbs sampler introduced in Andrieu et al. (2009, 2010). The resulting so-called Particle Gibbs Split-Merge sampler does not require the computation of a complex acceptance ratio and can be implemented using existing sequential Monte Carlo libraries. We investigate its performance experimentally on synthetic problems as well as on geolocation data. Our results show that for a given computational budget, the Particle Gibbs Split-Merge sampler empirically outperforms existing split merge methods. The code and instructions allowing to reproduce the experiments is available at https://github.com/aroth85/pgsm. |
| spellingShingle | Bouchard-Côté, A Doucet, A Roth, A Particle Gibbs split-merge sampling for Bayesian inference in mixture models |
| title | Particle Gibbs split-merge sampling for Bayesian inference in mixture models |
| title_full | Particle Gibbs split-merge sampling for Bayesian inference in mixture models |
| title_fullStr | Particle Gibbs split-merge sampling for Bayesian inference in mixture models |
| title_full_unstemmed | Particle Gibbs split-merge sampling for Bayesian inference in mixture models |
| title_short | Particle Gibbs split-merge sampling for Bayesian inference in mixture models |
| title_sort | particle gibbs split merge sampling for bayesian inference in mixture models |
| work_keys_str_mv | AT bouchardcotea particlegibbssplitmergesamplingforbayesianinferenceinmixturemodels AT douceta particlegibbssplitmergesamplingforbayesianinferenceinmixturemodels AT rotha particlegibbssplitmergesamplingforbayesianinferenceinmixturemodels |