A Unified Approach to Clustered and Non-stationary Linear Bandits
Abstract:
Non-stationary bandits and clustered bandits lift the restrictive assumptions in contextual bandits and provide solutions to many important real-world scenarios. Though they have been studied independently so far, we point out the essence in solving these two problems overlaps considerably. In this report, we unify these two strands of bandit research into a more general problem, i.e., contextual bandit in a clustered and non-stationary environment, and propose algorithms that can address the new challenges in this problem. Rigorous regret analysis and extensive empirical evaluations have been conducted to validate the proposed solutions.
Committee:
- Aidong Zhang, Committee Chair
- Hongning Wang, Advisor
- Haifeng Xu
- Sebastian Elbaum