@inproceedings{kumar-etal-2019-reinforcement,
title = "Reinforcement Learning based Curriculum Optimization for Neural Machine Translation",
author = "Kumar, Gaurav and
Foster, George and
Cherry, Colin and
Krikun, Maxim",
editor = "Burstein, Jill and
Doran, Christy and
Solorio, Thamar",
booktitle = "Proceedings of the 2019 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)",
month = jun,
year = "2019",
address = "Minneapolis, Minnesota",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/N19-1208",
doi = "10.18653/v1/N19-1208",
pages = "2054--2061",
abstract = "We consider the problem of making efficient use of heterogeneous training data in neural machine translation (NMT). Specifically, given a training dataset with a sentence-level feature such as noise, we seek an optimal curriculum, or order for presenting examples to the system during training. Our curriculum framework allows examples to appear an arbitrary number of times, and thus generalizes data weighting, filtering, and fine-tuning schemes. Rather than relying on prior knowledge to design a curriculum, we use reinforcement learning to learn one automatically, jointly with the NMT system, in the course of a single training run. We show that this approach can beat uniform baselines on Paracrawl and WMT English-to-French datasets by +3.4 and +1.3 BLEU respectively. Additionally, we match the performance of strong filtering baselines and hand-designed, state-of-the-art curricula.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kumar-etal-2019-reinforcement">
<titleInfo>
<title>Reinforcement Learning based Curriculum Optimization for Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Gaurav</namePart>
<namePart type="family">Kumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">George</namePart>
<namePart type="family">Foster</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Colin</namePart>
<namePart type="family">Cherry</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maxim</namePart>
<namePart type="family">Krikun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2019-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jill</namePart>
<namePart type="family">Burstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christy</namePart>
<namePart type="family">Doran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thamar</namePart>
<namePart type="family">Solorio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Minneapolis, Minnesota</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We consider the problem of making efficient use of heterogeneous training data in neural machine translation (NMT). Specifically, given a training dataset with a sentence-level feature such as noise, we seek an optimal curriculum, or order for presenting examples to the system during training. Our curriculum framework allows examples to appear an arbitrary number of times, and thus generalizes data weighting, filtering, and fine-tuning schemes. Rather than relying on prior knowledge to design a curriculum, we use reinforcement learning to learn one automatically, jointly with the NMT system, in the course of a single training run. We show that this approach can beat uniform baselines on Paracrawl and WMT English-to-French datasets by +3.4 and +1.3 BLEU respectively. Additionally, we match the performance of strong filtering baselines and hand-designed, state-of-the-art curricula.</abstract>
<identifier type="citekey">kumar-etal-2019-reinforcement</identifier>
<identifier type="doi">10.18653/v1/N19-1208</identifier>
<location>
<url>https://aclanthology.org/N19-1208</url>
</location>
<part>
<date>2019-06</date>
<extent unit="page">
<start>2054</start>
<end>2061</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Reinforcement Learning based Curriculum Optimization for Neural Machine Translation
%A Kumar, Gaurav
%A Foster, George
%A Cherry, Colin
%A Krikun, Maxim
%Y Burstein, Jill
%Y Doran, Christy
%Y Solorio, Thamar
%S Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
%D 2019
%8 June
%I Association for Computational Linguistics
%C Minneapolis, Minnesota
%F kumar-etal-2019-reinforcement
%X We consider the problem of making efficient use of heterogeneous training data in neural machine translation (NMT). Specifically, given a training dataset with a sentence-level feature such as noise, we seek an optimal curriculum, or order for presenting examples to the system during training. Our curriculum framework allows examples to appear an arbitrary number of times, and thus generalizes data weighting, filtering, and fine-tuning schemes. Rather than relying on prior knowledge to design a curriculum, we use reinforcement learning to learn one automatically, jointly with the NMT system, in the course of a single training run. We show that this approach can beat uniform baselines on Paracrawl and WMT English-to-French datasets by +3.4 and +1.3 BLEU respectively. Additionally, we match the performance of strong filtering baselines and hand-designed, state-of-the-art curricula.
%R 10.18653/v1/N19-1208
%U https://aclanthology.org/N19-1208
%U https://doi.org/10.18653/v1/N19-1208
%P 2054-2061
Markdown (Informal)
[Reinforcement Learning based Curriculum Optimization for Neural Machine Translation](https://aclanthology.org/N19-1208) (Kumar et al., NAACL 2019)
ACL
- Gaurav Kumar, George Foster, Colin Cherry, and Maxim Krikun. 2019. Reinforcement Learning based Curriculum Optimization for Neural Machine Translation. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 2054–2061, Minneapolis, Minnesota. Association for Computational Linguistics.