Thresholded linear bandits

Simple item page
dc.contributor.authorNguyen, Trang Thu
dc.contributor.supervisorMehta, Nishant
dc.date.accessioned2025-05-02T20:30:52Z
dc.date.available2025-05-02T20:30:52Z
dc.date.issued2025
dc.degree.departmentDepartment of Computer Science
dc.degree.levelMaster of Science MSc
dc.description.abstractThresholded linear bandits is a novel bandit problem that lies in the intersection of several important multiarmed bandit (MAB) variants, including active learning, structured bandits, and learning halfspaces. To achieve sublinear regret in the presence of exponentially many arms, one method is to exploit the structure of the reward function. However, the presence of an unknown threshold component makes previously known algorithms for structured bandits unsuitable. Moreover, the threshold introduces a discontinuity to the reward function, making the problem significantly more difficult. In this thesis, we study the union of axis-parallel halfspace variant of the thresholded linear bandits problem. We suggest an algorithm that achieves sublinear regret and provide theoretical guarantees on the performance of the algorithm
dc.description.scholarlevelGraduate
dc.identifier.urihttps://hdl.handle.net/1828/22112
dc.languageEnglisheng
dc.language.isoen
dc.rightsAvailable to the World Wide Web
dc.subjectmultiarmed bandits
dc.subjectmachine learning theory
dc.subjectthresholded bandits
dc.titleThresholded linear bandits
dc.typeThesis

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Nguyen_Trang_MSc_2025.pdf
Size:
318.73 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.62 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Electronic Theses and Dissertations (ETD)
Theses (Computer Science)
 
University of Victoria Libraries

We acknowledge and respect the Lək̓ʷəŋən (Songhees and Esquimalt) Peoples on whose territory the university stands, and the Lək̓ʷəŋən and W̱SÁNEĆ Peoples whose historical relationships with the land continue to this day.