<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="toledo-etal-2023-policy">
    <titleInfo>
        <title>Policy-based Reinforcement Learning for Generalisation in Interactive Text-based Environments</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Edan</namePart>
        <namePart type="family">Toledo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jan</namePart>
        <namePart type="family">Buys</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jonathan</namePart>
        <namePart type="family">Shock</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2023-05</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Andreas</namePart>
            <namePart type="family">Vlachos</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Isabelle</namePart>
            <namePart type="family">Augenstein</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Dubrovnik, Croatia</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Text-based environments enable RL agents to learn to converse and perform interactive tasks through natural language. However, previous RL approaches applied to text-based environments show poor performance when evaluated on unseen games. This paper investigates the improvement of generalisation performance through the simple switch from a value-based update method to a policy-based one, within text-based environments. We show that by replacing commonly used value-based methods with REINFORCE with baseline, a far more general agent is produced. The policy-based agent is evaluated on Coin Collector and Question Answering with interactive text (QAit), two text-based environments designed to test zero-shot performance. We see substantial improvements on a variety of zero-shot evaluation experiments, including tripling accuracy on various QAit benchmark configurations. The results indicate that policy-based RL has significantly better generalisation capabilities than value-based methods within such text-based environments, suggesting that RL agents could be applied to more complex natural language environments.</abstract>
    <identifier type="citekey">toledo-etal-2023-policy</identifier>
    <identifier type="doi">10.18653/v1/2023.eacl-main.88</identifier>
    <location>
        <url>https://aclanthology.org/2023.eacl-main.88</url>
    </location>
    <part>
        <date>2023-05</date>
        <extent unit="page">
            <start>1230</start>
            <end>1242</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Policy-based Reinforcement Learning for Generalisation in Interactive Text-based Environments
%A Toledo, Edan
%A Buys, Jan
%A Shock, Jonathan
%Y Vlachos, Andreas
%Y Augenstein, Isabelle
%S Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics
%D 2023
%8 May
%I Association for Computational Linguistics
%C Dubrovnik, Croatia
%F toledo-etal-2023-policy
%X Text-based environments enable RL agents to learn to converse and perform interactive tasks through natural language. However, previous RL approaches applied to text-based environments show poor performance when evaluated on unseen games. This paper investigates the improvement of generalisation performance through the simple switch from a value-based update method to a policy-based one, within text-based environments. We show that by replacing commonly used value-based methods with REINFORCE with baseline, a far more general agent is produced. The policy-based agent is evaluated on Coin Collector and Question Answering with interactive text (QAit), two text-based environments designed to test zero-shot performance. We see substantial improvements on a variety of zero-shot evaluation experiments, including tripling accuracy on various QAit benchmark configurations. The results indicate that policy-based RL has significantly better generalisation capabilities than value-based methods within such text-based environments, suggesting that RL agents could be applied to more complex natural language environments.
%R 10.18653/v1/2023.eacl-main.88
%U https://aclanthology.org/2023.eacl-main.88
%U https://doi.org/10.18653/v1/2023.eacl-main.88
%P 1230-1242

Markdown (Informal)

[Policy-based Reinforcement Learning for Generalisation in Interactive Text-based Environments](https://aclanthology.org/2023.eacl-main.88) (Toledo et al., EACL 2023)

• Policy-based Reinforcement Learning for Generalisation in Interactive Text-based Environments (Toledo et al., EACL 2023)

ACL

• Edan Toledo, Jan Buys, and Jonathan Shock. 2023. Policy-based Reinforcement Learning for Generalisation in Interactive Text-based Environments. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 1230–1242, Dubrovnik, Croatia. Association for Computational Linguistics.