@inproceedings{hashimoto-tsuruoka-2019-accelerated,
title = "Accelerated Reinforcement Learning for Sentence Generation by Vocabulary Prediction",
author = "Hashimoto, Kazuma and
Tsuruoka, Yoshimasa",
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/N19-1315",
doi = "10.18653/v1/N19-1315",
pages = "3115--3125",
abstract = "A major obstacle in reinforcement learning-based sentence generation is the large action space whose size is equal to the vocabulary size of the target-side language. To improve the efficiency of reinforcement learning, we present a novel approach for reducing the action space based on dynamic vocabulary prediction. Our method first predicts a fixed-size small vocabulary for each input to generate its target sentence. The input-specific vocabularies are then used at supervised and reinforcement learning steps, and also at test time. In our experiments on six machine translation and two image captioning datasets, our method achieves faster reinforcement learning ({\textasciitilde}2.7x faster) with less GPU memory ({\textasciitilde}2.3x less) than the full-vocabulary counterpart. We also show that our method more effectively receives rewards with fewer iterations of supervised pre-training.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hashimoto-tsuruoka-2019-accelerated">
<titleInfo>
<title>Accelerated Reinforcement Learning for Sentence Generation by Vocabulary Prediction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kazuma</namePart>
<namePart type="family">Hashimoto</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yoshimasa</namePart>
<namePart type="family">Tsuruoka</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>A major obstacle in reinforcement learning-based sentence generation is the large action space whose size is equal to the vocabulary size of the target-side language. To improve the efficiency of reinforcement learning, we present a novel approach for reducing the action space based on dynamic vocabulary prediction. Our method first predicts a fixed-size small vocabulary for each input to generate its target sentence. The input-specific vocabularies are then used at supervised and reinforcement learning steps, and also at test time. In our experiments on six machine translation and two image captioning datasets, our method achieves faster reinforcement learning (~2.7x faster) with less GPU memory (~2.3x less) than the full-vocabulary counterpart. We also show that our method more effectively receives rewards with fewer iterations of supervised pre-training.</abstract>
<identifier type="citekey">hashimoto-tsuruoka-2019-accelerated</identifier>
<identifier type="doi">10.18653/v1/N19-1315</identifier>
<location>
<url>https://aclanthology.org/N19-1315</url>
</location>
<part>
<date>2019-06</date>
<extent unit="page">
<start>3115</start>
<end>3125</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Accelerated Reinforcement Learning for Sentence Generation by Vocabulary Prediction
%A Hashimoto, Kazuma
%A Tsuruoka, Yoshimasa
%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 hashimoto-tsuruoka-2019-accelerated
%X A major obstacle in reinforcement learning-based sentence generation is the large action space whose size is equal to the vocabulary size of the target-side language. To improve the efficiency of reinforcement learning, we present a novel approach for reducing the action space based on dynamic vocabulary prediction. Our method first predicts a fixed-size small vocabulary for each input to generate its target sentence. The input-specific vocabularies are then used at supervised and reinforcement learning steps, and also at test time. In our experiments on six machine translation and two image captioning datasets, our method achieves faster reinforcement learning (~2.7x faster) with less GPU memory (~2.3x less) than the full-vocabulary counterpart. We also show that our method more effectively receives rewards with fewer iterations of supervised pre-training.
%R 10.18653/v1/N19-1315
%U https://aclanthology.org/N19-1315
%U https://doi.org/10.18653/v1/N19-1315
%P 3115-3125
Markdown (Informal)
[Accelerated Reinforcement Learning for Sentence Generation by Vocabulary Prediction](https://aclanthology.org/N19-1315) (Hashimoto & Tsuruoka, NAACL 2019)
ACL