2026-09-19T03:27:46Zhttps://ecommons.cornell.edu/server/oai/requestoai:ecommons.cornell.edu:1813/590502026-05-15T19:51:00Zcom_1813_35col_1813_47
Chen, Bangrui
Frazier, Peter
Topaloglu, Huseyin
Joachims, Thorsten
2018-10-03T19:27:23Z
2018-10-03T19:27:23Z
2017-12-30
ProQuest Submission ID: 10605
ProQuest Publication ID: 10680541
https://hdl.handle.net/1813/59050
https://doi.org/10.7298/X4251GCQ
10474153
In this thesis, we study adaptive preference learning, in which a machine learning system learns users' preferences from feedback while simultaneously using these learned preferences to help them find preferred items. We study three different types of user feedback in three application setting: cardinal feedback with application in information filtering systems, ordinal feedback with application in personalized content recommender systems, and attribute feedback with application in review aggregators. We connect these settings respectively to existing work on classical multi-armed bandits, dueling bandits, and incentivizing exploration. For each type of feedback and application setting, we provide an algorithm and a theoretical analysis bounding its regret. We demonstrate through numerical experiments that our algorithms outperform existing benchmarks.
en_US
Statistics
Operations research
Computer science
adaptive preference learning
bandit feedback
dueling bandits
incentivizing exploration
information filtering
multi-armed bandits
Adaptive Preference Learning With Bandit Feedback: Information Filtering, Dueling Bandits and Incentivizing Exploration
dissertation or thesis
application/pdf
Operations Research
Cornell University
Doctor of Philosophy
Ph. D., Operations Research
https://hdl.handle.net/1813/59810
Publication
Cornell Theses and Dissertations
Chen, Bangrui
5893a6ea-7af3-41d7-abc6-04bcd26ab5df
open.access
open.access
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="088cc75e-8398-4cd7-a279-0f7975fb4e33">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Language>en_US</Language>
<Title>Adaptive Preference Learning With Bandit Feedback: Information Filtering, Dueling Bandits and Incentivizing Exploration</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2017-12-30</PublicationDate>
<DOI>https://doi.org/10.7298/X4251GCQ</DOI>
<Authors>
<Author>
<DisplayName>Chen, Bangrui</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<OrgUnit />
</Publisher>
</Publishers>
<Keyword>Statistics</Keyword>
<Keyword>Operations research</Keyword>
<Keyword>Computer science</Keyword>
<Keyword>adaptive preference learning</Keyword>
<Keyword>bandit feedback</Keyword>
<Keyword>dueling bandits</Keyword>
<Keyword>incentivizing exploration</Keyword>
<Keyword>information filtering</Keyword>
<Keyword>multi-armed bandits</Keyword>
<Abstract>In this thesis, we study adaptive preference learning, in which a machine learning system learns users' preferences from feedback while simultaneously using these learned preferences to help them find preferred items. We study three different types of user feedback in three application setting: cardinal feedback with application in information filtering systems, ordinal feedback with application in personalized content recommender systems, and attribute feedback with application in review aggregators. We connect these settings respectively to existing work on classical multi-armed bandits, dueling bandits, and incentivizing exploration. For each type of feedback and application setting, we provide an algorithm and a theoretical analysis bounding its regret. We demonstrate through numerical experiments that our algorithms outperform existing benchmarks.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>