Learning OT constraint rankings using a maximum entropy model

Sharon Goldwater; Mark Johnson

Learning OT constraint rankings using a maximum entropy model

Sharon Goldwater, Mark Johnson

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

Abstract

A weakness of standard Optimality Theory is its inability to account for grammars with free variation. We describe here the Maximum Entropy model, a general statistical model, and show how it can be applied in a constraint-based linguistic framework to model and learn grammars with free variation, as well as categorical grammars. We report the results of using the MaxEnt model for learning two different grammars: one with variation, and one without. Our results are as good as those of a previous probabilistic version of OT, the Gradual Learning Algorithm (Boersma, 1997), and we argue that our model is more general and mathematically well-motivated.

Original language	English
Title of host publication	Proceedings of the Stockholm Workshop on Variation within Optimality Theory
Editors	Jennifer Spenader, Anders Eriksson, Östen Dahl
Place of Publication	Stockholm
Publisher	Stockholm University
Pages	111-120
Number of pages	10
Publication status	Published - 2003
Externally published	Yes
Event	Stockholm Workshop on Variation within Optimality Theory - Stockholm University, Stockholm, Sweden Duration: 26 Apr 2003 → 27 Apr 2003

Workshop

Workshop	Stockholm Workshop on Variation within Optimality Theory
Country/Territory	Sweden
City	Stockholm
Period	26/04/03 → 27/04/03

Cite this

@inproceedings{ac36bf71ddbb47b2ae6ae567e2f30bef,

title = "Learning OT constraint rankings using a maximum entropy model",

abstract = "A weakness of standard Optimality Theory is its inability to account for grammars with free variation. We describe here the Maximum Entropy model, a general statistical model, and show how it can be applied in a constraint-based linguistic framework to model and learn grammars with free variation, as well as categorical grammars. We report the results of using the MaxEnt model for learning two different grammars: one with variation, and one without. Our results are as good as those of a previous probabilistic version of OT, the Gradual Learning Algorithm (Boersma, 1997), and we argue that our model is more general and mathematically well-motivated.",

author = "Sharon Goldwater and Mark Johnson",

year = "2003",

language = "English",

pages = "111--120",

editor = "Jennifer Spenader and Anders Eriksson and {\"O}sten Dahl",

booktitle = "Proceedings of the Stockholm Workshop on Variation within Optimality Theory",

publisher = "Stockholm University",

note = "Stockholm Workshop on Variation within Optimality Theory ; Conference date: 26-04-2003 Through 27-04-2003",

}

Learning OT constraint rankings using a maximum entropy model. / Goldwater, Sharon; Johnson, Mark.
Proceedings of the Stockholm Workshop on Variation within Optimality Theory. ed. / Jennifer Spenader; Anders Eriksson; Östen Dahl. Stockholm: Stockholm University, 2003. p. 111-120.

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

TY - GEN

T1 - Learning OT constraint rankings using a maximum entropy model

AU - Goldwater, Sharon

AU - Johnson, Mark

PY - 2003

Y1 - 2003

N2 - A weakness of standard Optimality Theory is its inability to account for grammars with free variation. We describe here the Maximum Entropy model, a general statistical model, and show how it can be applied in a constraint-based linguistic framework to model and learn grammars with free variation, as well as categorical grammars. We report the results of using the MaxEnt model for learning two different grammars: one with variation, and one without. Our results are as good as those of a previous probabilistic version of OT, the Gradual Learning Algorithm (Boersma, 1997), and we argue that our model is more general and mathematically well-motivated.

AB - A weakness of standard Optimality Theory is its inability to account for grammars with free variation. We describe here the Maximum Entropy model, a general statistical model, and show how it can be applied in a constraint-based linguistic framework to model and learn grammars with free variation, as well as categorical grammars. We report the results of using the MaxEnt model for learning two different grammars: one with variation, and one without. Our results are as good as those of a previous probabilistic version of OT, the Gradual Learning Algorithm (Boersma, 1997), and we argue that our model is more general and mathematically well-motivated.

M3 - Conference proceeding contribution

SP - 111

EP - 120

BT - Proceedings of the Stockholm Workshop on Variation within Optimality Theory

A2 - Spenader, Jennifer

A2 - Eriksson, Anders

A2 - Dahl, Östen

PB - Stockholm University

CY - Stockholm

T2 - Stockholm Workshop on Variation within Optimality Theory

Y2 - 26 April 2003 through 27 April 2003

ER -

Learning OT constraint rankings using a maximum entropy model

Abstract

Workshop

Fingerprint

Cite this