Learning to actively learn neural machine translation

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

Abstract

Traditional active learning (AL) methods for machine translation (MT) rely on heuristics. However, these heuristics are limited when the characteristics of the MT problem change due to e.g. the language pair or the amount of the initial bitext. In this paper, we present a framework to learn sentence selection strategies for neural MT. We train the AL query strategy using a high-resource language-pair based on AL simulations, and then transfer it to the lowresource language-pair of interest. The learned query strategy capitalizes on the shared characteristics between the language pairs to make an effective use of the AL budget. Our experiments on three language-pairs confirms that our method is more effective than strong heuristic-based methods in various conditions, including cold-start and warm-start as well as small and extremely small data conditions.
Original languageEnglish
Title of host publicationCoNLL 2018 - The 22nd Conference on Computational Natural Language Learning - Proceedings of the Conference
Subtitle of host publicationOctober 31 - November 1, 2018 Brussels, Belgium
EditorsMiikka Silfverberg
Place of PublicationStroudsburg PA USA
PublisherThe Association for Computational Linguistics
Pages334-344
Number of pages11
ISBN (Electronic)9781948087728
Publication statusPublished - 2018
EventConference on Natural Language Learning 2018 - Brussels, Belgium
Duration: 31 Oct 20181 Nov 2018
https://www.conll.org/2018

Conference

ConferenceConference on Natural Language Learning 2018
Abbreviated titleCoNLL 2018
CountryBelgium
CityBrussels
Period31/10/181/11/18
Internet address

Cite this

Liu, M., Buntine, W., & Haffari, G. (2018). Learning to actively learn neural machine translation. In M. Silfverberg (Ed.), CoNLL 2018 - The 22nd Conference on Computational Natural Language Learning - Proceedings of the Conference: October 31 - November 1, 2018 Brussels, Belgium (pp. 334-344). Stroudsburg PA USA: The Association for Computational Linguistics.
Liu, Ming ; Buntine, Wray ; Haffari, Gholamreza. / Learning to actively learn neural machine translation. CoNLL 2018 - The 22nd Conference on Computational Natural Language Learning - Proceedings of the Conference: October 31 - November 1, 2018 Brussels, Belgium. editor / Miikka Silfverberg. Stroudsburg PA USA : The Association for Computational Linguistics, 2018. pp. 334-344
@inproceedings{977dd36d45f042c09dae60b3901ed3b1,
title = "Learning to actively learn neural machine translation",
abstract = "Traditional active learning (AL) methods for machine translation (MT) rely on heuristics. However, these heuristics are limited when the characteristics of the MT problem change due to e.g. the language pair or the amount of the initial bitext. In this paper, we present a framework to learn sentence selection strategies for neural MT. We train the AL query strategy using a high-resource language-pair based on AL simulations, and then transfer it to the lowresource language-pair of interest. The learned query strategy capitalizes on the shared characteristics between the language pairs to make an effective use of the AL budget. Our experiments on three language-pairs confirms that our method is more effective than strong heuristic-based methods in various conditions, including cold-start and warm-start as well as small and extremely small data conditions.",
author = "Ming Liu and Wray Buntine and Gholamreza Haffari",
year = "2018",
language = "English",
pages = "334--344",
editor = "Silfverberg, {Miikka }",
booktitle = "CoNLL 2018 - The 22nd Conference on Computational Natural Language Learning - Proceedings of the Conference",
publisher = "The Association for Computational Linguistics",

}

Liu, M, Buntine, W & Haffari, G 2018, Learning to actively learn neural machine translation. in M Silfverberg (ed.), CoNLL 2018 - The 22nd Conference on Computational Natural Language Learning - Proceedings of the Conference: October 31 - November 1, 2018 Brussels, Belgium. The Association for Computational Linguistics, Stroudsburg PA USA, pp. 334-344, Conference on Natural Language Learning 2018, Brussels, Belgium, 31/10/18.

Learning to actively learn neural machine translation. / Liu, Ming; Buntine, Wray; Haffari, Gholamreza.

CoNLL 2018 - The 22nd Conference on Computational Natural Language Learning - Proceedings of the Conference: October 31 - November 1, 2018 Brussels, Belgium. ed. / Miikka Silfverberg. Stroudsburg PA USA : The Association for Computational Linguistics, 2018. p. 334-344.

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

TY - GEN

T1 - Learning to actively learn neural machine translation

AU - Liu, Ming

AU - Buntine, Wray

AU - Haffari, Gholamreza

PY - 2018

Y1 - 2018

N2 - Traditional active learning (AL) methods for machine translation (MT) rely on heuristics. However, these heuristics are limited when the characteristics of the MT problem change due to e.g. the language pair or the amount of the initial bitext. In this paper, we present a framework to learn sentence selection strategies for neural MT. We train the AL query strategy using a high-resource language-pair based on AL simulations, and then transfer it to the lowresource language-pair of interest. The learned query strategy capitalizes on the shared characteristics between the language pairs to make an effective use of the AL budget. Our experiments on three language-pairs confirms that our method is more effective than strong heuristic-based methods in various conditions, including cold-start and warm-start as well as small and extremely small data conditions.

AB - Traditional active learning (AL) methods for machine translation (MT) rely on heuristics. However, these heuristics are limited when the characteristics of the MT problem change due to e.g. the language pair or the amount of the initial bitext. In this paper, we present a framework to learn sentence selection strategies for neural MT. We train the AL query strategy using a high-resource language-pair based on AL simulations, and then transfer it to the lowresource language-pair of interest. The learned query strategy capitalizes on the shared characteristics between the language pairs to make an effective use of the AL budget. Our experiments on three language-pairs confirms that our method is more effective than strong heuristic-based methods in various conditions, including cold-start and warm-start as well as small and extremely small data conditions.

M3 - Conference Paper

SP - 334

EP - 344

BT - CoNLL 2018 - The 22nd Conference on Computational Natural Language Learning - Proceedings of the Conference

A2 - Silfverberg, Miikka

PB - The Association for Computational Linguistics

CY - Stroudsburg PA USA

ER -

Liu M, Buntine W, Haffari G. Learning to actively learn neural machine translation. In Silfverberg M, editor, CoNLL 2018 - The 22nd Conference on Computational Natural Language Learning - Proceedings of the Conference: October 31 - November 1, 2018 Brussels, Belgium. Stroudsburg PA USA: The Association for Computational Linguistics. 2018. p. 334-344