<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kocmi-bojar-2018-trivial">
    <titleInfo>
        <title>Trivial Transfer Learning for Low-Resource Neural Machine Translation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Tom</namePart>
        <namePart type="family">Kocmi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ondřej</namePart>
        <namePart type="family">Bojar</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2018-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Third Conference on Machine Translation: Research Papers</title>
        </titleInfo>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Brussels, Belgium</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Transfer learning has been proven as an effective technique for neural machine translation under low-resource conditions. Existing methods require a common target language, language relatedness, or specific training tricks and regimes. We present a simple transfer learning method, where we first train a “parent” model for a high-resource language pair and then continue the training on a low-resource pair only by replacing the training corpus. This “child” model performs significantly better than the baseline trained for low-resource pair only. We are the first to show this for targeting different languages, and we observe the improvements even for unrelated languages with different alphabets.</abstract>
    <identifier type="citekey">kocmi-bojar-2018-trivial</identifier>
    <identifier type="doi">10.18653/v1/W18-6325</identifier>
    <location>
        <url>https://aclanthology.org/W18-6325</url>
    </location>
    <part>
        <date>2018-10</date>
        <extent unit="page">
            <start>244</start>
            <end>252</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Trivial Transfer Learning for Low-Resource Neural Machine Translation
%A Kocmi, Tom
%A Bojar, Ondřej
%S Proceedings of the Third Conference on Machine Translation: Research Papers
%D 2018
%8 October
%I Association for Computational Linguistics
%C Brussels, Belgium
%F kocmi-bojar-2018-trivial
%X Transfer learning has been proven as an effective technique for neural machine translation under low-resource conditions. Existing methods require a common target language, language relatedness, or specific training tricks and regimes. We present a simple transfer learning method, where we first train a “parent” model for a high-resource language pair and then continue the training on a low-resource pair only by replacing the training corpus. This “child” model performs significantly better than the baseline trained for low-resource pair only. We are the first to show this for targeting different languages, and we observe the improvements even for unrelated languages with different alphabets.
%R 10.18653/v1/W18-6325
%U https://aclanthology.org/W18-6325
%U https://doi.org/10.18653/v1/W18-6325
%P 244-252

Markdown (Informal)

[Trivial Transfer Learning for Low-Resource Neural Machine Translation](https://aclanthology.org/W18-6325) (Kocmi & Bojar, WMT 2018)

• Trivial Transfer Learning for Low-Resource Neural Machine Translation (Kocmi & Bojar, WMT 2018)

ACL

• Tom Kocmi and Ondřej Bojar. 2018. Trivial Transfer Learning for Low-Resource Neural Machine Translation. In Proceedings of the Third Conference on Machine Translation: Research Papers, pages 244–252, Brussels, Belgium. Association for Computational Linguistics.