<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hua-etal-2022-numerical">
    <titleInfo>
        <title>Numerical Optimizations for Weighted Low-rank Estimation on Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Ting</namePart>
        <namePart type="family">Hua</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yen-Chang</namePart>
        <namePart type="family">Hsu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Felicity</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Qian</namePart>
        <namePart type="family">Lou</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yilin</namePart>
        <namePart type="family">Shen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hongxia</namePart>
        <namePart type="family">Jin</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2022-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yoav</namePart>
            <namePart type="family">Goldberg</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Zornitsa</namePart>
            <namePart type="family">Kozareva</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yue</namePart>
            <namePart type="family">Zhang</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Singular value decomposition (SVD) is one of the most popular compression methods that approximate a target matrix with smaller matrices. However, standard SVD treats the parameters within the matrix with equal importance, which is a simple but unrealistic assumption. The parameters of a trained neural network model may affect the task performance unevenly, which suggests non-equal importance among the parameters. Compared to SVD, the decomposition method aware of parameter importance is the more practical choice in real cases. Unlike standard SVD, weighed value decomposition is a non-convex optimization problem that lacks a closed-form solution. We systematically investigated multiple optimization strategies to tackle the problem and examined our method by compressing Transformer-based language models.Further, we designed a metric to predict when the SVD may introduce a significant performance drop, for which our method can be a rescue strategy.The extensive evaluations demonstrate that our method can perform better than current SOTA methods in compressing Transformer-based language models.</abstract>
    <identifier type="citekey">hua-etal-2022-numerical</identifier>
    <identifier type="doi">10.18653/v1/2022.emnlp-main.91</identifier>
    <location>
        <url>https://aclanthology.org/2022.emnlp-main.91</url>
    </location>
    <part>
        <date>2022-12</date>
        <extent unit="page">
            <start>1404</start>
            <end>1416</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Numerical Optimizations for Weighted Low-rank Estimation on Language Models
%A Hua, Ting
%A Hsu, Yen-Chang
%A Wang, Felicity
%A Lou, Qian
%A Shen, Yilin
%A Jin, Hongxia
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F hua-etal-2022-numerical
%X Singular value decomposition (SVD) is one of the most popular compression methods that approximate a target matrix with smaller matrices. However, standard SVD treats the parameters within the matrix with equal importance, which is a simple but unrealistic assumption. The parameters of a trained neural network model may affect the task performance unevenly, which suggests non-equal importance among the parameters. Compared to SVD, the decomposition method aware of parameter importance is the more practical choice in real cases. Unlike standard SVD, weighed value decomposition is a non-convex optimization problem that lacks a closed-form solution. We systematically investigated multiple optimization strategies to tackle the problem and examined our method by compressing Transformer-based language models.Further, we designed a metric to predict when the SVD may introduce a significant performance drop, for which our method can be a rescue strategy.The extensive evaluations demonstrate that our method can perform better than current SOTA methods in compressing Transformer-based language models.
%R 10.18653/v1/2022.emnlp-main.91
%U https://aclanthology.org/2022.emnlp-main.91
%U https://doi.org/10.18653/v1/2022.emnlp-main.91
%P 1404-1416

Markdown (Informal)

[Numerical Optimizations for Weighted Low-rank Estimation on Language Models](https://aclanthology.org/2022.emnlp-main.91) (Hua et al., EMNLP 2022)

• Numerical Optimizations for Weighted Low-rank Estimation on Language Models (Hua et al., EMNLP 2022)

ACL

• Ting Hua, Yen-Chang Hsu, Felicity Wang, Qian Lou, Yilin Shen, and Hongxia Jin. 2022. Numerical Optimizations for Weighted Low-rank Estimation on Language Models. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 1404–1416, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.