<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lalor-yu-2020-dynamic">
    <titleInfo>
        <title>Dynamic Data Selection for Curriculum Learning via Ability Estimation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">John</namePart>
        <namePart type="given">P</namePart>
        <namePart type="family">Lalor</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hong</namePart>
        <namePart type="family">Yu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2020-nov</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Findings of the Association for Computational Linguistics: EMNLP 2020</title>
        </titleInfo>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Online</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Curriculum learning methods typically rely on heuristics to estimate the difficulty of training examples or the ability of the model. In this work, we propose replacing difficulty heuristics with learned difficulty parameters. We also propose Dynamic Data selection for Curriculum Learning via Ability Estimation (DDaCLAE), a strategy that probes model ability at each training epoch to select the best training examples at that point. We show that models using learned difficulty and/or ability outperform heuristic-based curriculum learning models on the GLUE classification tasks.</abstract>
    <identifier type="citekey">lalor-yu-2020-dynamic</identifier>
    <identifier type="doi">10.18653/v1/2020.findings-emnlp.48</identifier>
    <location>
        <url>https://aclanthology.org/2020.findings-emnlp.48</url>
    </location>
    <part>
        <date>2020-nov</date>
        <extent unit="page">
            <start>545</start>
            <end>555</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Dynamic Data Selection for Curriculum Learning via Ability Estimation
%A Lalor, John P.
%A Yu, Hong
%S Findings of the Association for Computational Linguistics: EMNLP 2020
%D 2020
%8 nov
%I Association for Computational Linguistics
%C Online
%F lalor-yu-2020-dynamic
%X Curriculum learning methods typically rely on heuristics to estimate the difficulty of training examples or the ability of the model. In this work, we propose replacing difficulty heuristics with learned difficulty parameters. We also propose Dynamic Data selection for Curriculum Learning via Ability Estimation (DDaCLAE), a strategy that probes model ability at each training epoch to select the best training examples at that point. We show that models using learned difficulty and/or ability outperform heuristic-based curriculum learning models on the GLUE classification tasks.
%R 10.18653/v1/2020.findings-emnlp.48
%U https://aclanthology.org/2020.findings-emnlp.48
%U https://doi.org/10.18653/v1/2020.findings-emnlp.48
%P 545-555

Markdown (Informal)

[Dynamic Data Selection for Curriculum Learning via Ability Estimation](https://aclanthology.org/2020.findings-emnlp.48) (Lalor & Yu, Findings 2020)

• Dynamic Data Selection for Curriculum Learning via Ability Estimation (Lalor & Yu, Findings 2020)

ACL

• John P. Lalor and Hong Yu. 2020. Dynamic Data Selection for Curriculum Learning via Ability Estimation. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 545–555, Online. Association for Computational Linguistics.