Adaptor grammars: A framework for specifying compositional nonparametric Bayesian models

Mark Johnson; Thomas L. Griffiths; Sharon Goldwater

Adaptor grammars: A framework for specifying compositional nonparametric Bayesian models

Mark Johnson^*, Thomas L. Griffiths, Sharon Goldwater

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

162 Citations (Scopus)

Abstract

This paper introduces adaptor grammars, a class of probabilistic models of language that generalize probabilistic context-free grammars (PCFGs). Adaptor grammars augment the probabilistic rules of PCFGs with "adaptors" that can induce dependencies among successive uses. With a particular choice of adaptor, based on the Pitman-Yor process, nonparametric Bayesian models of language using Dirichlet processes and hierarchical Dirichlet processes can be written as simple grammars. We present a general-purpose inference algorithm for adaptor grammars, making it easy to define and use such models, and illustrate how several existing nonparametric Bayesian models can be expressed within this framework.

Original language	English
Title of host publication	Advances in Neural Information Processing Systems 19 - Proceedings of the 2006 Conference
Editors	Bernhard Scholkopf, John Platt, Thomas Hofmann
Place of Publication	Cambridge, MA
Publisher	MIT Press
Pages	641-648
Number of pages	8
ISBN (Print)	9780262195683
Publication status	Published - 2007
Externally published	Yes
Event	20th Annual Conference on Neural Information Processing Systems, NIPS 2006 - Vancouver, BC, Canada Duration: 4 Dec 2006 → 7 Dec 2006

Other

Other	20th Annual Conference on Neural Information Processing Systems, NIPS 2006
Country/Territory	Canada
City	Vancouver, BC
Period	4/12/06 → 7/12/06

Cite this

@inproceedings{c9acd8fc79c242e882399e9d6bb6d5fd,

title = "Adaptor grammars: A framework for specifying compositional nonparametric Bayesian models",

abstract = "This paper introduces adaptor grammars, a class of probabilistic models of language that generalize probabilistic context-free grammars (PCFGs). Adaptor grammars augment the probabilistic rules of PCFGs with {"}adaptors{"} that can induce dependencies among successive uses. With a particular choice of adaptor, based on the Pitman-Yor process, nonparametric Bayesian models of language using Dirichlet processes and hierarchical Dirichlet processes can be written as simple grammars. We present a general-purpose inference algorithm for adaptor grammars, making it easy to define and use such models, and illustrate how several existing nonparametric Bayesian models can be expressed within this framework.",

author = "Mark Johnson and Griffiths, \{Thomas L.\} and Sharon Goldwater",

year = "2007",

language = "English",

isbn = "9780262195683",

pages = "641--648",

editor = "Bernhard Scholkopf and John Platt and Thomas Hofmann",

booktitle = "Advances in Neural Information Processing Systems 19 - Proceedings of the 2006 Conference",

publisher = "MIT Press",

note = "20th Annual Conference on Neural Information Processing Systems, NIPS 2006 ; Conference date: 04-12-2006 Through 07-12-2006",

}

Johnson, M, Griffiths, TL & Goldwater, S 2007, Adaptor grammars: A framework for specifying compositional nonparametric Bayesian models. in B Scholkopf, J Platt & T Hofmann (eds), Advances in Neural Information Processing Systems 19 - Proceedings of the 2006 Conference. MIT Press, Cambridge, MA, pp. 641-648, 20th Annual Conference on Neural Information Processing Systems, NIPS 2006, Vancouver, BC, Canada, 4/12/06.

Adaptor grammars: A framework for specifying compositional nonparametric Bayesian models. / Johnson, Mark; Griffiths, Thomas L.; Goldwater, Sharon.
Advances in Neural Information Processing Systems 19 - Proceedings of the 2006 Conference. ed. / Bernhard Scholkopf; John Platt; Thomas Hofmann. Cambridge, MA: MIT Press, 2007. p. 641-648.

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - Adaptor grammars

T2 - 20th Annual Conference on Neural Information Processing Systems, NIPS 2006

AU - Johnson, Mark

AU - Griffiths, Thomas L.

AU - Goldwater, Sharon

PY - 2007

Y1 - 2007

N2 - This paper introduces adaptor grammars, a class of probabilistic models of language that generalize probabilistic context-free grammars (PCFGs). Adaptor grammars augment the probabilistic rules of PCFGs with "adaptors" that can induce dependencies among successive uses. With a particular choice of adaptor, based on the Pitman-Yor process, nonparametric Bayesian models of language using Dirichlet processes and hierarchical Dirichlet processes can be written as simple grammars. We present a general-purpose inference algorithm for adaptor grammars, making it easy to define and use such models, and illustrate how several existing nonparametric Bayesian models can be expressed within this framework.

AB - This paper introduces adaptor grammars, a class of probabilistic models of language that generalize probabilistic context-free grammars (PCFGs). Adaptor grammars augment the probabilistic rules of PCFGs with "adaptors" that can induce dependencies among successive uses. With a particular choice of adaptor, based on the Pitman-Yor process, nonparametric Bayesian models of language using Dirichlet processes and hierarchical Dirichlet processes can be written as simple grammars. We present a general-purpose inference algorithm for adaptor grammars, making it easy to define and use such models, and illustrate how several existing nonparametric Bayesian models can be expressed within this framework.

UR - https://www.scopus.com/pages/publications/67349214856

M3 - Conference proceeding contribution

AN - SCOPUS:67349214856

SN - 9780262195683

SP - 641

EP - 648

BT - Advances in Neural Information Processing Systems 19 - Proceedings of the 2006 Conference

A2 - Scholkopf, Bernhard

A2 - Platt, John

A2 - Hofmann, Thomas

PB - MIT Press

CY - Cambridge, MA

Y2 - 4 December 2006 through 7 December 2006

ER -

Adaptor grammars: A framework for specifying compositional nonparametric Bayesian models

Abstract

Other

Other files and links

Fingerprint

Cite this