A note on the implementation of hierarchical Dirichlet processes

Phil Blunsom*, Sharon Goldwater, Trevor Cohn, Mark Johnson

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

12 Citations (Scopus)

Abstract

The implementation of collapsed Gibbs samplers for non-parametric Bayesian models is non-trivial, requiring considerable book-keeping. Goldwater et al. (2006a) presented an approximation which significantly reduces the storage and computation overhead, but we show here that their formulation was incorrect and, even after correction, is grossly inaccurate. We present an alternative formulation which is exact and can be computed easily. However this approach does not work for hierarchical models, for which case we present an efficient data structure which has a better space complexity than the naive approach.

Original languageEnglish
Title of host publicationProceedings of the ACL-IJCNLP 2009 Conference Short Papers
Place of PublicationSingapore
PublisherAssociation for Computational Linguistics (ACL)
Pages337-340
Number of pages4
ISBN (Print)9781617382581
Publication statusPublished - 2009
Externally publishedYes
EventJoint Conference of the 47th Annual Meeting of the Association for Computational Linguistics and 4th International Joint Conference on Natural Language Processing of the AFNLP, ACL-IJCNLP 2009 - Suntec, Singapore
Duration: 2 Aug 20097 Aug 2009

Other

OtherJoint Conference of the 47th Annual Meeting of the Association for Computational Linguistics and 4th International Joint Conference on Natural Language Processing of the AFNLP, ACL-IJCNLP 2009
CountrySingapore
CitySuntec
Period2/08/097/08/09

    Fingerprint

Cite this

Blunsom, P., Goldwater, S., Cohn, T., & Johnson, M. (2009). A note on the implementation of hierarchical Dirichlet processes. In Proceedings of the ACL-IJCNLP 2009 Conference Short Papers (pp. 337-340). Singapore: Association for Computational Linguistics (ACL).