@inproceedings{komiya-shinnou-2018-investigating,
title = "Investigating Effective Parameters for Fine-tuning of Word Embeddings Using Only a Small Corpus",
author = "Komiya, Kanako and
Shinnou, Hiroyuki",
editor = "Haffari, Reza and
Cherry, Colin and
Foster, George and
Khadivi, Shahram and
Salehi, Bahar",
booktitle = "Proceedings of the Workshop on Deep Learning Approaches for Low-Resource {NLP}",
month = jul,
year = "2018",
address = "Melbourne",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/W18-3408",
doi = "10.18653/v1/W18-3408",
pages = "60--67",
abstract = "Fine-tuning is a popular method to achieve better performance when only a small target corpus is available. However, it requires tuning of a number of metaparameters and thus it might carry risk of adverse effect when inappropriate metaparameters are used. Therefore, we investigate effective parameters for fine-tuning when only a small target corpus is available. In the current study, we target at improving Japanese word embeddings created from a huge corpus. First, we demonstrate that even the word embeddings created from the huge corpus are affected by domain shift. After that, we investigate effective parameters for fine-tuning of the word embeddings using a small target corpus. We used perplexity of a language model obtained from a Long Short-Term Memory network to assess the word embeddings input into the network. The experiments revealed that fine-tuning sometimes give adverse effect when only a small target corpus is used and batch size is the most important parameter for fine-tuning. In addition, we confirmed that effect of fine-tuning is higher when size of a target corpus was larger.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="komiya-shinnou-2018-investigating">
<titleInfo>
<title>Investigating Effective Parameters for Fine-tuning of Word Embeddings Using Only a Small Corpus</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kanako</namePart>
<namePart type="family">Komiya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hiroyuki</namePart>
<namePart type="family">Shinnou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2018-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Workshop on Deep Learning Approaches for Low-Resource NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Reza</namePart>
<namePart type="family">Haffari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Colin</namePart>
<namePart type="family">Cherry</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">George</namePart>
<namePart type="family">Foster</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shahram</namePart>
<namePart type="family">Khadivi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bahar</namePart>
<namePart type="family">Salehi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Melbourne</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Fine-tuning is a popular method to achieve better performance when only a small target corpus is available. However, it requires tuning of a number of metaparameters and thus it might carry risk of adverse effect when inappropriate metaparameters are used. Therefore, we investigate effective parameters for fine-tuning when only a small target corpus is available. In the current study, we target at improving Japanese word embeddings created from a huge corpus. First, we demonstrate that even the word embeddings created from the huge corpus are affected by domain shift. After that, we investigate effective parameters for fine-tuning of the word embeddings using a small target corpus. We used perplexity of a language model obtained from a Long Short-Term Memory network to assess the word embeddings input into the network. The experiments revealed that fine-tuning sometimes give adverse effect when only a small target corpus is used and batch size is the most important parameter for fine-tuning. In addition, we confirmed that effect of fine-tuning is higher when size of a target corpus was larger.</abstract>
<identifier type="citekey">komiya-shinnou-2018-investigating</identifier>
<identifier type="doi">10.18653/v1/W18-3408</identifier>
<location>
<url>https://aclanthology.org/W18-3408</url>
</location>
<part>
<date>2018-07</date>
<extent unit="page">
<start>60</start>
<end>67</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Investigating Effective Parameters for Fine-tuning of Word Embeddings Using Only a Small Corpus
%A Komiya, Kanako
%A Shinnou, Hiroyuki
%Y Haffari, Reza
%Y Cherry, Colin
%Y Foster, George
%Y Khadivi, Shahram
%Y Salehi, Bahar
%S Proceedings of the Workshop on Deep Learning Approaches for Low-Resource NLP
%D 2018
%8 July
%I Association for Computational Linguistics
%C Melbourne
%F komiya-shinnou-2018-investigating
%X Fine-tuning is a popular method to achieve better performance when only a small target corpus is available. However, it requires tuning of a number of metaparameters and thus it might carry risk of adverse effect when inappropriate metaparameters are used. Therefore, we investigate effective parameters for fine-tuning when only a small target corpus is available. In the current study, we target at improving Japanese word embeddings created from a huge corpus. First, we demonstrate that even the word embeddings created from the huge corpus are affected by domain shift. After that, we investigate effective parameters for fine-tuning of the word embeddings using a small target corpus. We used perplexity of a language model obtained from a Long Short-Term Memory network to assess the word embeddings input into the network. The experiments revealed that fine-tuning sometimes give adverse effect when only a small target corpus is used and batch size is the most important parameter for fine-tuning. In addition, we confirmed that effect of fine-tuning is higher when size of a target corpus was larger.
%R 10.18653/v1/W18-3408
%U https://aclanthology.org/W18-3408
%U https://doi.org/10.18653/v1/W18-3408
%P 60-67
Markdown (Informal)
[Investigating Effective Parameters for Fine-tuning of Word Embeddings Using Only a Small Corpus](https://aclanthology.org/W18-3408) (Komiya & Shinnou, ACL 2018)
ACL