@inproceedings{budzianowski-etal-2017-sub,
title = "Sub-domain Modelling for Dialogue Management with Hierarchical Reinforcement Learning",
author = "Budzianowski, Pawe{\l} and
Ultes, Stefan and
Su, Pei-Hao and
Mrk{\v{s}}i{\'c}, Nikola and
Wen, Tsung-Hsien and
Casanueva, I{\~n}igo and
Rojas-Barahona, Lina M. and
Ga{\v{s}}i{\'c}, Milica",
editor = "Jokinen, Kristiina and
Stede, Manfred and
DeVault, David and
Louis, Annie",
booktitle = "Proceedings of the 18th Annual {SIG}dial Meeting on Discourse and Dialogue",
month = aug,
year = "2017",
address = {Saarbr{\"u}cken, Germany},
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/W17-5512",
doi = "10.18653/v1/W17-5512",
pages = "86--92",
abstract = "Human conversation is inherently complex, often spanning many different topics/domains. This makes policy learning for dialogue systems very challenging. Standard flat reinforcement learning methods do not provide an efficient framework for modelling such dialogues. In this paper, we focus on the under-explored problem of multi-domain dialogue management. First, we propose a new method for hierarchical reinforcement learning using the option framework. Next, we show that the proposed architecture learns faster and arrives at a better policy than the existing flat ones do. Moreover, we show how pretrained policies can be adapted to more complex systems with an additional set of new actions. In doing that, we show that our approach has the potential to facilitate policy optimisation for more sophisticated multi-domain dialogue systems.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="budzianowski-etal-2017-sub">
<titleInfo>
<title>Sub-domain Modelling for Dialogue Management with Hierarchical Reinforcement Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Paweł</namePart>
<namePart type="family">Budzianowski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stefan</namePart>
<namePart type="family">Ultes</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pei-Hao</namePart>
<namePart type="family">Su</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikola</namePart>
<namePart type="family">Mrkšić</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tsung-Hsien</namePart>
<namePart type="family">Wen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Iñigo</namePart>
<namePart type="family">Casanueva</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>
<name type="personal">
<namePart type="given">Milica</namePart>
<namePart type="family">Gašić</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2017-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kristiina</namePart>
<namePart type="family">Jokinen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Manfred</namePart>
<namePart type="family">Stede</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">DeVault</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Annie</namePart>
<namePart type="family">Louis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Saarbrücken, Germany</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Human conversation is inherently complex, often spanning many different topics/domains. This makes policy learning for dialogue systems very challenging. Standard flat reinforcement learning methods do not provide an efficient framework for modelling such dialogues. In this paper, we focus on the under-explored problem of multi-domain dialogue management. First, we propose a new method for hierarchical reinforcement learning using the option framework. Next, we show that the proposed architecture learns faster and arrives at a better policy than the existing flat ones do. Moreover, we show how pretrained policies can be adapted to more complex systems with an additional set of new actions. In doing that, we show that our approach has the potential to facilitate policy optimisation for more sophisticated multi-domain dialogue systems.</abstract>
<identifier type="citekey">budzianowski-etal-2017-sub</identifier>
<identifier type="doi">10.18653/v1/W17-5512</identifier>
<location>
<url>https://aclanthology.org/W17-5512</url>
</location>
<part>
<date>2017-08</date>
<extent unit="page">
<start>86</start>
<end>92</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Sub-domain Modelling for Dialogue Management with Hierarchical Reinforcement Learning
%A Budzianowski, Paweł
%A Ultes, Stefan
%A Su, Pei-Hao
%A Mrkšić, Nikola
%A Wen, Tsung-Hsien
%A Casanueva, Iñigo
%A Rojas-Barahona, Lina M.
%A Gašić, Milica
%Y Jokinen, Kristiina
%Y Stede, Manfred
%Y DeVault, David
%Y Louis, Annie
%S Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue
%D 2017
%8 August
%I Association for Computational Linguistics
%C Saarbrücken, Germany
%F budzianowski-etal-2017-sub
%X Human conversation is inherently complex, often spanning many different topics/domains. This makes policy learning for dialogue systems very challenging. Standard flat reinforcement learning methods do not provide an efficient framework for modelling such dialogues. In this paper, we focus on the under-explored problem of multi-domain dialogue management. First, we propose a new method for hierarchical reinforcement learning using the option framework. Next, we show that the proposed architecture learns faster and arrives at a better policy than the existing flat ones do. Moreover, we show how pretrained policies can be adapted to more complex systems with an additional set of new actions. In doing that, we show that our approach has the potential to facilitate policy optimisation for more sophisticated multi-domain dialogue systems.
%R 10.18653/v1/W17-5512
%U https://aclanthology.org/W17-5512
%U https://doi.org/10.18653/v1/W17-5512
%P 86-92
Markdown (Informal)
[Sub-domain Modelling for Dialogue Management with Hierarchical Reinforcement Learning](https://aclanthology.org/W17-5512) (Budzianowski et al., SIGDIAL 2017)
ACL
- Paweł Budzianowski, Stefan Ultes, Pei-Hao Su, Nikola Mrkšić, Tsung-Hsien Wen, Iñigo Casanueva, Lina M. Rojas-Barahona, and Milica Gašić. 2017. Sub-domain Modelling for Dialogue Management with Hierarchical Reinforcement Learning. In Proceedings of the 18th Annual SIGdial Meeting on Discourse and Dialogue, pages 86–92, Saarbrücken, Germany. Association for Computational Linguistics.
