Active learning for deep semantic parsing

Long  Duong; Hadi  Afshar; Dominique  Estival; Glen  Pink; Philip Cohen; Mark Johnson

Active learning for deep semantic parsing

Long Duong, Hadi Afshar, Dominique Estival, Glen Pink, Philip Cohen, Mark Johnson

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

Abstract

Semantic parsing requires training data that is expensive and slow to collect. We apply active learning to both traditional and “overnight” data collection approaches. We show that it is possible to obtain good training hyperparameters from seed data which is only a small fraction of the full dataset. We show that uncertainty sampling based on least confidence score is competitive in traditional data collection but not applicable for overnight collection. We propose several active learning strategies for overnight data collection and show that different example selection strategies per domain perform best.

Original language	English
Title of host publication	ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics - Proceedings of the Conference, Vol. 2 (Short Papers) July 15
Subtitle of host publication	July 15 - 20, 2018 Melbourne, Australia
Editors	Iryna Gurevych, Yusuke Miyao
Place of Publication	Stroudsburg PA USA
Publisher	Association for Computational Linguistics (ACL)
Pages	43-48
Number of pages	6
Volume	2
ISBN (Electronic)	9781948087346
Publication status	Published - 2018
Externally published	Yes
Event	Annual Meeting of the Association of Computational Linguistics 2018 - Melbourne, Australia Duration: 15 Jul 2018 → 20 Jul 2018 Conference number: 56th https://aclanthology.info/events/acl-2018

Conference

Conference	Annual Meeting of the Association of Computational Linguistics 2018
Abbreviated title	ACL 2018
Country/Territory	Australia
City	Melbourne
Period	15/07/18 → 20/07/18
Internet address	https://aclanthology.info/events/acl-2018

Keywords

semantic parsing
machine learning
Active learning

Access to Document

https://aclweb.org/anthology/P18-2008

Cite this

Duong, L., Afshar, H., Estival, D., Pink, G., Cohen, P., & Johnson, M. (2018). Active learning for deep semantic parsing. In I. Gurevych, & Y. Miyao (Eds.), ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics - Proceedings of the Conference, Vol. 2 (Short Papers) July 15 : July 15 - 20, 2018 Melbourne, Australia (Vol. 2, pp. 43-48). Association for Computational Linguistics (ACL). https://aclweb.org/anthology/P18-2008

Duong, Long ; Afshar, Hadi ; Estival, Dominique et al. / Active learning for deep semantic parsing. ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics - Proceedings of the Conference, Vol. 2 (Short Papers) July 15 : July 15 - 20, 2018 Melbourne, Australia. editor / Iryna Gurevych ; Yusuke Miyao. Vol. 2 Stroudsburg PA USA : Association for Computational Linguistics (ACL), 2018. pp. 43-48

@inproceedings{33aac172fc4e424484b82dd9972070e7,

title = "Active learning for deep semantic parsing",

abstract = "Semantic parsing requires training data that is expensive and slow to collect. We apply active learning to both traditional and “overnight” data collection approaches. We show that it is possible to obtain good training hyperparameters from seed data which is only a small fraction of the full dataset. We show that uncertainty sampling based on least confidence score is competitive in traditional data collection but not applicable for overnight collection. We propose several active learning strategies for overnight data collection and show that different example selection strategies per domain perform best.",

keywords = "semantic parsing, machine learning, Active learning",

author = "Long Duong and Hadi Afshar and Dominique Estival and Glen Pink and Philip Cohen and Mark Johnson",

year = "2018",

language = "English",

volume = "2",

pages = "43--48",

editor = "Iryna Gurevych and Yusuke Miyao",

booktitle = "ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics - Proceedings of the Conference, Vol. 2 (Short Papers) July 15",

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

note = "Annual Meeting of the Association of Computational Linguistics 2018, ACL 2018 ; Conference date: 15-07-2018 Through 20-07-2018",

url = "https://aclanthology.info/events/acl-2018",

}

Duong, L, Afshar, H, Estival, D, Pink, G, Cohen, P & Johnson, M 2018, Active learning for deep semantic parsing. in I Gurevych & Y Miyao (eds), ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics - Proceedings of the Conference, Vol. 2 (Short Papers) July 15 : July 15 - 20, 2018 Melbourne, Australia. vol. 2, Association for Computational Linguistics (ACL), Stroudsburg PA USA, pp. 43-48, Annual Meeting of the Association of Computational Linguistics 2018, Melbourne, Victoria, Australia, 15/07/18. <https://aclweb.org/anthology/P18-2008>

Active learning for deep semantic parsing. / Duong, Long ; Afshar, Hadi ; Estival, Dominique et al.

ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics - Proceedings of the Conference, Vol. 2 (Short Papers) July 15 : July 15 - 20, 2018 Melbourne, Australia. ed. / Iryna Gurevych; Yusuke Miyao. Vol. 2 Stroudsburg PA USA : Association for Computational Linguistics (ACL), 2018. p. 43-48.

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

TY - GEN

T1 - Active learning for deep semantic parsing

AU - Duong, Long

AU - Afshar, Hadi

AU - Estival, Dominique

AU - Pink, Glen

AU - Cohen, Philip

AU - Johnson, Mark

N1 - Conference code: 56th

PY - 2018

Y1 - 2018

N2 - Semantic parsing requires training data that is expensive and slow to collect. We apply active learning to both traditional and “overnight” data collection approaches. We show that it is possible to obtain good training hyperparameters from seed data which is only a small fraction of the full dataset. We show that uncertainty sampling based on least confidence score is competitive in traditional data collection but not applicable for overnight collection. We propose several active learning strategies for overnight data collection and show that different example selection strategies per domain perform best.

AB - Semantic parsing requires training data that is expensive and slow to collect. We apply active learning to both traditional and “overnight” data collection approaches. We show that it is possible to obtain good training hyperparameters from seed data which is only a small fraction of the full dataset. We show that uncertainty sampling based on least confidence score is competitive in traditional data collection but not applicable for overnight collection. We propose several active learning strategies for overnight data collection and show that different example selection strategies per domain perform best.

KW - semantic parsing

KW - machine learning

KW - Active learning

M3 - Conference Paper

VL - 2

SP - 43

EP - 48

BT - ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics - Proceedings of the Conference, Vol. 2 (Short Papers) July 15

A2 - Gurevych, Iryna

A2 - Miyao, Yusuke

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg PA USA

T2 - Annual Meeting of the Association of Computational Linguistics 2018

Y2 - 15 July 2018 through 20 July 2018

ER -

Duong L, Afshar H, Estival D, Pink G, Cohen P, Johnson M. Active learning for deep semantic parsing. In Gurevych I, Miyao Y, editors, ACL 2018 - The 56th Annual Meeting of the Association for Computational Linguistics - Proceedings of the Conference, Vol. 2 (Short Papers) July 15 : July 15 - 20, 2018 Melbourne, Australia. Vol. 2. Stroudsburg PA USA: Association for Computational Linguistics (ACL). 2018. p. 43-48