Reinforcement learning for spoken dialogue systems using off-policy natural gradient method
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F12%3A10194751" target="_blank" >RIV/00216208:11320/12:10194751 - isvavai.cz</a>
Result on the web
<a href="http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6424161" target="_blank" >http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6424161</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Reinforcement learning for spoken dialogue systems using off-policy natural gradient method
Original language description
Reinforcement learning methods have been successfully used to optimise dialogue strategies in statistical dialogue systems. Typically, reinforcement techniques learn on-policy i.e., the dialogue strategy is updated online while the system is interactingwith a user. An alternative to this approach is off-policy reinforcement learning, which estimates an optimal dialogue strategy offline from a fixed corpus of previously collected dialogues. This paper proposes a novel off-policy reinforcement learning method based on natural policy gradients and importance sampling. The algorithm is evaluated on a spoken dialogue system in the tourist information domain. The experiments indicate that the proposed method learns a dialogue strategy, which significantly outperforms the baseline handcrafted dialogue policy
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
IN - Informatics
OECD FORD branch
—
Result continuities
Project
<a href="/en/project/LK11221" target="_blank" >LK11221: Development of statistical methods for spoken dalogue systems</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)
Others
Publication year
2012
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
IEEE SLT '12: Proc. IEEE Spoken Language Technology Workshop
ISBN
978-1-4673-5126-3
ISSN
—
e-ISSN
—
Number of pages
6
Pages from-to
7-12
Publisher name
IEEE
Place of publication
Miami, FL, USA
Event location
Miami, FL, USA
Event date
Dec 2, 2012
Type of event by nationality
CST - Celostátní akce
UT code for WoS article
—