Reducing grounded learning tasks to grammatical inference

Benjamin Börschinger; Bevan K. Jones; Mark Johnson

Reducing grounded learning tasks to grammatical inference

Benjamin Börschinger^*, Bevan K. Jones, Mark Johnson

^*Corresponding author for this work

School of Computing

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

26 Citations (Scopus)

Abstract

It is often assumed that 'grounded' learning tasks are beyond the scope of grammatical inference techniques. In this paper, we show that the grounded task of learning a semantic parser from ambiguous training data as discussed in Kim and Mooney (2010) can be reduced to a Probabilistic Context-Free Grammar learning task in a way that gives state of the art results. We further show that additionally letting our model learn the language's canonical word order improves its performance and leads to the highest semantic parsing f-scores previously reported in the literature.

Original language	English
Title of host publication	EMNLP 2011 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
Place of Publication	Edinburgh, UK
Publisher	Association for Computational Linguistics (ACL)
Pages	1416-1425
Number of pages	10
ISBN (Print)	1937284115, 9781937284114
Publication status	Published - 2011
Event	Conference on Empirical Methods in Natural Language Processing, EMNLP 2011 - Edinburgh, United Kingdom Duration: 27 Jul 2011 → 31 Jul 2011

Other

Other	Conference on Empirical Methods in Natural Language Processing, EMNLP 2011
Country/Territory	United Kingdom
City	Edinburgh
Period	27/07/11 → 31/07/11

Cite this

@inproceedings{13d3e814f44b495ca8180f50677d7ca8,

title = "Reducing grounded learning tasks to grammatical inference",

abstract = "It is often assumed that 'grounded' learning tasks are beyond the scope of grammatical inference techniques. In this paper, we show that the grounded task of learning a semantic parser from ambiguous training data as discussed in Kim and Mooney (2010) can be reduced to a Probabilistic Context-Free Grammar learning task in a way that gives state of the art results. We further show that additionally letting our model learn the language's canonical word order improves its performance and leads to the highest semantic parsing f-scores previously reported in the literature.",

author = "Benjamin B{\"o}rschinger and Jones, {Bevan K.} and Mark Johnson",

year = "2011",

language = "English",

isbn = "1937284115",

pages = "1416--1425",

booktitle = "EMNLP 2011 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference",

publisher = "Association for Computational Linguistics (ACL)",

note = "Conference on Empirical Methods in Natural Language Processing, EMNLP 2011 ; Conference date: 27-07-2011 Through 31-07-2011",

}

Börschinger, B, Jones, BK & Johnson, M 2011, Reducing grounded learning tasks to grammatical inference. in EMNLP 2011 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. Association for Computational Linguistics (ACL), Edinburgh, UK, pp. 1416-1425, Conference on Empirical Methods in Natural Language Processing, EMNLP 2011, Edinburgh, United Kingdom, 27/07/11.

Reducing grounded learning tasks to grammatical inference. / Börschinger, Benjamin; Jones, Bevan K.; Johnson, Mark.
EMNLP 2011 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference. Edinburgh, UK: Association for Computational Linguistics (ACL), 2011. p. 1416-1425.

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

TY - GEN

T1 - Reducing grounded learning tasks to grammatical inference

AU - Börschinger, Benjamin

AU - Jones, Bevan K.

AU - Johnson, Mark

PY - 2011

Y1 - 2011

N2 - It is often assumed that 'grounded' learning tasks are beyond the scope of grammatical inference techniques. In this paper, we show that the grounded task of learning a semantic parser from ambiguous training data as discussed in Kim and Mooney (2010) can be reduced to a Probabilistic Context-Free Grammar learning task in a way that gives state of the art results. We further show that additionally letting our model learn the language's canonical word order improves its performance and leads to the highest semantic parsing f-scores previously reported in the literature.

AB - It is often assumed that 'grounded' learning tasks are beyond the scope of grammatical inference techniques. In this paper, we show that the grounded task of learning a semantic parser from ambiguous training data as discussed in Kim and Mooney (2010) can be reduced to a Probabilistic Context-Free Grammar learning task in a way that gives state of the art results. We further show that additionally letting our model learn the language's canonical word order improves its performance and leads to the highest semantic parsing f-scores previously reported in the literature.

UR - http://www.scopus.com/inward/record.url?scp=80053254449&partnerID=8YFLogxK

M3 - Conference proceeding contribution

AN - SCOPUS:80053254449

SN - 1937284115

SN - 9781937284114

SP - 1416

EP - 1425

BT - EMNLP 2011 - Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference

PB - Association for Computational Linguistics (ACL)

CY - Edinburgh, UK

T2 - Conference on Empirical Methods in Natural Language Processing, EMNLP 2011

Y2 - 27 July 2011 through 31 July 2011

ER -

Reducing grounded learning tasks to grammatical inference

Abstract

Other

Other files and links

Fingerprint

Cite this