@inproceedings{cordier-etal-2023-shot,
title = "Few-Shot Structured Policy Learning for Multi-Domain and Multi-Task Dialogues",
author = "Cordier, Thibault and
Urvoy, Tanguy and
Lef{\`e}vre, Fabrice and
Rojas Barahona, Lina M.",
editor = "Vlachos, Andreas and
Augenstein, Isabelle",
booktitle = "Findings of the Association for Computational Linguistics: EACL 2023",
month = may,
year = "2023",
address = "Dubrovnik, Croatia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-eacl.32",
doi = "10.18653/v1/2023.findings-eacl.32",
pages = "432--441",
abstract = "Reinforcement learning has been widely adopted to model dialogue managers in task-oriented dialogues. However, the user simulator provided by state-of-the-art dialogue frameworks are only rough approximations of human behaviour. The ability to learn from a small number of human interactions is hence crucial, especially on multi-domain and multi-task environments where the action space is large. We therefore propose to use structured policies to improve sample efficiency when learning on these kinds of environments. We also evaluate the impact of learning from human vs simulated experts. Among the different levels of structure that we tested, the graph neural networks (GNNs) show a remarkable superiority by reaching a success rate above 80{\%} with only 50 dialogues when learning from simulated experts. They also show superiority when learning from human experts, although a performance drop was observed. We therefore suggest to concentrate future research efforts on bridging the gap between human data, simulators and automatic evaluators in dialogue frameworks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cordier-etal-2023-shot">
<titleInfo>
<title>Few-Shot Structured Policy Learning for Multi-Domain and Multi-Task Dialogues</title>
</titleInfo>
<name type="personal">
<namePart type="given">Thibault</namePart>
<namePart type="family">Cordier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanguy</namePart>
<namePart type="family">Urvoy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabrice</namePart>
<namePart type="family">Lefèvre</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lina</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Rojas Barahona</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>Findings of the Association for Computational Linguistics: EACL 2023</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>Reinforcement learning has been widely adopted to model dialogue managers in task-oriented dialogues. However, the user simulator provided by state-of-the-art dialogue frameworks are only rough approximations of human behaviour. The ability to learn from a small number of human interactions is hence crucial, especially on multi-domain and multi-task environments where the action space is large. We therefore propose to use structured policies to improve sample efficiency when learning on these kinds of environments. We also evaluate the impact of learning from human vs simulated experts. Among the different levels of structure that we tested, the graph neural networks (GNNs) show a remarkable superiority by reaching a success rate above 80% with only 50 dialogues when learning from simulated experts. They also show superiority when learning from human experts, although a performance drop was observed. We therefore suggest to concentrate future research efforts on bridging the gap between human data, simulators and automatic evaluators in dialogue frameworks.</abstract>
<identifier type="citekey">cordier-etal-2023-shot</identifier>
<identifier type="doi">10.18653/v1/2023.findings-eacl.32</identifier>
<location>
<url>https://aclanthology.org/2023.findings-eacl.32</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>432</start>
<end>441</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Few-Shot Structured Policy Learning for Multi-Domain and Multi-Task Dialogues
%A Cordier, Thibault
%A Urvoy, Tanguy
%A Lefèvre, Fabrice
%A Rojas Barahona, Lina M.
%Y Vlachos, Andreas
%Y Augenstein, Isabelle
%S Findings of the Association for Computational Linguistics: EACL 2023
%D 2023
%8 May
%I Association for Computational Linguistics
%C Dubrovnik, Croatia
%F cordier-etal-2023-shot
%X Reinforcement learning has been widely adopted to model dialogue managers in task-oriented dialogues. However, the user simulator provided by state-of-the-art dialogue frameworks are only rough approximations of human behaviour. The ability to learn from a small number of human interactions is hence crucial, especially on multi-domain and multi-task environments where the action space is large. We therefore propose to use structured policies to improve sample efficiency when learning on these kinds of environments. We also evaluate the impact of learning from human vs simulated experts. Among the different levels of structure that we tested, the graph neural networks (GNNs) show a remarkable superiority by reaching a success rate above 80% with only 50 dialogues when learning from simulated experts. They also show superiority when learning from human experts, although a performance drop was observed. We therefore suggest to concentrate future research efforts on bridging the gap between human data, simulators and automatic evaluators in dialogue frameworks.
%R 10.18653/v1/2023.findings-eacl.32
%U https://aclanthology.org/2023.findings-eacl.32
%U https://doi.org/10.18653/v1/2023.findings-eacl.32
%P 432-441
Markdown (Informal)
[Few-Shot Structured Policy Learning for Multi-Domain and Multi-Task Dialogues](https://aclanthology.org/2023.findings-eacl.32) (Cordier et al., Findings 2023)
ACL