@inproceedings{argueta-chiang-2017-decoding,
title = "Decoding with Finite-State Transducers on {GPU}s",
author = "Argueta, Arturo and
Chiang, David",
editor = "Lapata, Mirella and
Blunsom, Phil and
Koller, Alexander",
booktitle = "Proceedings of the 15th Conference of the {E}uropean Chapter of the Association for Computational Linguistics: Volume 1, Long Papers",
month = apr,
year = "2017",
address = "Valencia, Spain",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/E17-1098",
pages = "1044--1052",
abstract = "Weighted finite automata and transducers (including hidden Markov models and conditional random fields) are widely used in natural language processing (NLP) to perform tasks such as morphological analysis, part-of-speech tagging, chunking, named entity recognition, speech recognition, and others. Parallelizing finite state algorithms on graphics processing units (GPUs) would benefit many areas of NLP. Although researchers have implemented GPU versions of basic graph algorithms, no work, to our knowledge, has been done on GPU algorithms for weighted finite automata. We introduce a GPU implementation of the Viterbi and forward-backward algorithm, achieving speedups of up to 4x over our serial implementations running on different computer architectures and 3335x over widely used tools such as OpenFST.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="argueta-chiang-2017-decoding">
<titleInfo>
<title>Decoding with Finite-State Transducers on GPUs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Arturo</namePart>
<namePart type="family">Argueta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Chiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2017-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mirella</namePart>
<namePart type="family">Lapata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Phil</namePart>
<namePart type="family">Blunsom</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Koller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Valencia, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Weighted finite automata and transducers (including hidden Markov models and conditional random fields) are widely used in natural language processing (NLP) to perform tasks such as morphological analysis, part-of-speech tagging, chunking, named entity recognition, speech recognition, and others. Parallelizing finite state algorithms on graphics processing units (GPUs) would benefit many areas of NLP. Although researchers have implemented GPU versions of basic graph algorithms, no work, to our knowledge, has been done on GPU algorithms for weighted finite automata. We introduce a GPU implementation of the Viterbi and forward-backward algorithm, achieving speedups of up to 4x over our serial implementations running on different computer architectures and 3335x over widely used tools such as OpenFST.</abstract>
<identifier type="citekey">argueta-chiang-2017-decoding</identifier>
<location>
<url>https://aclanthology.org/E17-1098</url>
</location>
<part>
<date>2017-04</date>
<extent unit="page">
<start>1044</start>
<end>1052</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Decoding with Finite-State Transducers on GPUs
%A Argueta, Arturo
%A Chiang, David
%Y Lapata, Mirella
%Y Blunsom, Phil
%Y Koller, Alexander
%S Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers
%D 2017
%8 April
%I Association for Computational Linguistics
%C Valencia, Spain
%F argueta-chiang-2017-decoding
%X Weighted finite automata and transducers (including hidden Markov models and conditional random fields) are widely used in natural language processing (NLP) to perform tasks such as morphological analysis, part-of-speech tagging, chunking, named entity recognition, speech recognition, and others. Parallelizing finite state algorithms on graphics processing units (GPUs) would benefit many areas of NLP. Although researchers have implemented GPU versions of basic graph algorithms, no work, to our knowledge, has been done on GPU algorithms for weighted finite automata. We introduce a GPU implementation of the Viterbi and forward-backward algorithm, achieving speedups of up to 4x over our serial implementations running on different computer architectures and 3335x over widely used tools such as OpenFST.
%U https://aclanthology.org/E17-1098
%P 1044-1052
Markdown (Informal)
[Decoding with Finite-State Transducers on GPUs](https://aclanthology.org/E17-1098) (Argueta & Chiang, EACL 2017)
ACL
- Arturo Argueta and David Chiang. 2017. Decoding with Finite-State Transducers on GPUs. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 1044–1052, Valencia, Spain. Association for Computational Linguistics.