Parametrized Stochastic Multi-armed Bandits with Binary Rewards

Jiang, Chong

Permalink

https://hdl.handle.net/2142/18352 Copy

Description

Title

Parametrized Stochastic Multi-armed Bandits with Binary Rewards

Author(s)

Jiang, Chong

Issue Date

2011-01-14T22:47:13Z

Director of Research (if dissertation) or Advisor (if thesis)

Srikant, Rayadurgam

Department of Study

Electrical & Computer Eng

Discipline

Electrical & Computer Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• multi-armed bandits
• machine learning

Abstract

In this thesis, we consider the problem of multi-armed bandits with a large number of correlated arms. We assume that the arms have Bernoulli distributed rewards, independent across arms and across time, where the probabilities of success are parametrized by known attribute vectors for each arm, as well as an unknown preference vector. For this model, we seek an algorithm with a total regret that is sub-linear in time and independent of the number of arms. We present such an algorithm, which we call the Three-phase Algorithm, and analyze its performance. We show an upper bound on the total regret which applies uniformly in time. The asymptotics of this bound show that for any $f \in \omega(\log(T))$, the total regret can be made to be $O(f(T))$, independent of the number of arms.

Graduation Semester

2010-12

Permalink

http://hdl.handle.net/2142/18352

Copyright and License Information

Copyright 2010 Chong Jiang

Owning Collections