2026-09-19T15:38:06Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/403252022-01-13T07:54:29Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Nicholas Roy.
Doshi, Finale (Finale P.)
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2008-02-27T20:38:45Z
2008-02-27T20:38:45Z
2007
2007
http://hdl.handle.net/1721.1/40325
191957537
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2007.
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Includes bibliographical references (p. 118-122).
Partially Observable Markov Decision Processes (POMDPs) have succeeded in many planning domains because they can optimally trade between actions that will increase an agent's knowledge about its environment and actions that will increase an agent's reward. However, POMDPs are defined with a large number of parameters which are difficult to specify from domain knowledge, and gathering enough data to specify the parameters a priori may be expensive. This work develops several efficient algorithms for learning the POMDP parameters online and demonstrates them on dialog manager for a robotic wheelchair. In particular, we show how a combination of specialized queries ("meta-actions") can enable us to create a robust dialog manager that avoids the pitfalls in other POMDP-learning approaches. The dialog manager's ability to reason about its uncertainty -- and take advantage of low-risk opportunities to reduce that uncertainty -- leads to more robust policy learning.
by Final Doshi.
S.M.
122 p.
eng
Massachusetts Institute of Technology
M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.
http://dspace.mit.edu/handle/1721.1/7582
Electrical Engineering and Computer Science.
Efficient model learning for dialog management
Thesis
application/pdf
false
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="6e1ab93b-d912-4b74-9037-dcf7194b4d63">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Language>eng</Language>
<Title>Efficient model learning for dialog management</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2007</PublicationDate>
<Authors>
<Author>
<DisplayName>Doshi, Finale (Finale P.)</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://dspace.mit.edu/handle/1721.1/7582</License>
<Keyword>Electrical Engineering and Computer Science.</Keyword>
<Abstract>Partially Observable Markov Decision Processes (POMDPs) have succeeded in many planning domains because they can optimally trade between actions that will increase an agent's knowledge about its environment and actions that will increase an agent's reward. However, POMDPs are defined with a large number of parameters which are difficult to specify from domain knowledge, and gathering enough data to specify the parameters a priori may be expensive. This work develops several efficient algorithms for learning the POMDP parameters online and demonstrates them on dialog manager for a robotic wheelchair. In particular, we show how a combination of specialized queries ("meta-actions") can enable us to create a robust dialog manager that avoids the pitfalls in other POMDP-learning approaches. The dialog manager's ability to reason about its uncertainty -- and take advantage of low-risk opportunities to reduce that uncertainty -- leads to more robust policy learning.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>