@inproceedings{sankar-etal-2019-transferable,
title = "Transferable Neural Projection Representations",
author = "Sankar, Chinnadhurai and
Ravi, Sujith and
Kozareva, Zornitsa",
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/N19-1339",
doi = "10.18653/v1/N19-1339",
pages = "3355--3360",
abstract = "Neural word representations are at the core of many state-of-the-art natural language processing models. A widely used approach is to pre-train, store and look up word or character embedding matrices. While useful, such representations occupy huge memory making it hard to deploy on-device and often do not generalize to unknown words due to vocabulary pruning. In this paper, we propose a skip-gram based architecture coupled with Locality-Sensitive Hashing (LSH) projections to learn efficient dynamically computable representations. Our model does not need to store lookup tables as representations are computed on-the-fly and require low memory footprint. The representations can be trained in an unsupervised fashion and can be easily transferred to other NLP tasks. For qualitative evaluation, we analyze the nearest neighbors of the word representations and discover semantically similar words even with misspellings. For quantitative evaluation, we plug our transferable projections into a simple LSTM and run it on multiple NLP tasks and show how our transferable projections achieve better performance compared to prior work.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sankar-etal-2019-transferable">
<titleInfo>
<title>Transferable Neural Projection Representations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chinnadhurai</namePart>
<namePart type="family">Sankar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sujith</namePart>
<namePart type="family">Ravi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</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>Neural word representations are at the core of many state-of-the-art natural language processing models. A widely used approach is to pre-train, store and look up word or character embedding matrices. While useful, such representations occupy huge memory making it hard to deploy on-device and often do not generalize to unknown words due to vocabulary pruning. In this paper, we propose a skip-gram based architecture coupled with Locality-Sensitive Hashing (LSH) projections to learn efficient dynamically computable representations. Our model does not need to store lookup tables as representations are computed on-the-fly and require low memory footprint. The representations can be trained in an unsupervised fashion and can be easily transferred to other NLP tasks. For qualitative evaluation, we analyze the nearest neighbors of the word representations and discover semantically similar words even with misspellings. For quantitative evaluation, we plug our transferable projections into a simple LSTM and run it on multiple NLP tasks and show how our transferable projections achieve better performance compared to prior work.</abstract>
<identifier type="citekey">sankar-etal-2019-transferable</identifier>
<identifier type="doi">10.18653/v1/N19-1339</identifier>
<location>
<url>https://aclanthology.org/N19-1339</url>
</location>
<part>
<date>2019-06</date>
<extent unit="page">
<start>3355</start>
<end>3360</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Transferable Neural Projection Representations
%A Sankar, Chinnadhurai
%A Ravi, Sujith
%A Kozareva, Zornitsa
%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 sankar-etal-2019-transferable
%X Neural word representations are at the core of many state-of-the-art natural language processing models. A widely used approach is to pre-train, store and look up word or character embedding matrices. While useful, such representations occupy huge memory making it hard to deploy on-device and often do not generalize to unknown words due to vocabulary pruning. In this paper, we propose a skip-gram based architecture coupled with Locality-Sensitive Hashing (LSH) projections to learn efficient dynamically computable representations. Our model does not need to store lookup tables as representations are computed on-the-fly and require low memory footprint. The representations can be trained in an unsupervised fashion and can be easily transferred to other NLP tasks. For qualitative evaluation, we analyze the nearest neighbors of the word representations and discover semantically similar words even with misspellings. For quantitative evaluation, we plug our transferable projections into a simple LSTM and run it on multiple NLP tasks and show how our transferable projections achieve better performance compared to prior work.
%R 10.18653/v1/N19-1339
%U https://aclanthology.org/N19-1339
%U https://doi.org/10.18653/v1/N19-1339
%P 3355-3360
Markdown (Informal)
[Transferable Neural Projection Representations](https://aclanthology.org/N19-1339) (Sankar et al., NAACL 2019)
ACL
- Chinnadhurai Sankar, Sujith Ravi, and Zornitsa Kozareva. 2019. Transferable Neural Projection Representations. 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 3355–3360, Minneapolis, Minnesota. Association for Computational Linguistics.