English

Optimal Adaptive Learning in Uncontrolled Restless Bandit Problems

Optimization and Control 2015-01-30 v3 Machine Learning

Abstract

In this paper we consider the problem of learning the optimal policy for uncontrolled restless bandit problems. In an uncontrolled restless bandit problem, there is a finite set of arms, each of which when pulled yields a positive reward. There is a player who sequentially selects one of the arms at each time step. The goal of the player is to maximize its undiscounted reward over a time horizon T. The reward process of each arm is a finite state Markov chain, whose transition probabilities are unknown by the player. State transitions of each arm is independent of the selection of the player. We propose a learning algorithm with logarithmic regret uniformly over time with respect to the optimal finite horizon policy. Our results extend the optimal adaptive learning of MDPs to POMDPs.

Keywords

Cite

@article{arxiv.1107.4042,
  title  = {Optimal Adaptive Learning in Uncontrolled Restless Bandit Problems},
  author = {Cem Tekin and Mingyan Liu},
  journal= {arXiv preprint arXiv:1107.4042},
  year   = {2015}
}
R2 v1 2026-06-21T18:39:33.207Z