English

Regret Bounds for Restless Markov Bandits

Machine Learning 2012-10-23 v1 Optimization and Control Machine Learning

Abstract

We consider the restless Markov bandit problem, in which the state of each arm evolves according to a Markov process independently of the learner's actions. We suggest an algorithm that after TT steps achieves O~(T)\tilde{O}(\sqrt{T}) regret with respect to the best policy that knows the distributions of all arms. No assumptions on the Markov chains are made except that they are irreducible. In addition, we show that index-based policies are necessarily suboptimal for the considered problem.

Keywords

Cite

@article{arxiv.1209.2693,
  title  = {Regret Bounds for Restless Markov Bandits},
  author = {Ronald Ortner and Daniil Ryabko and Peter Auer and Rémi Munos},
  journal= {arXiv preprint arXiv:1209.2693},
  year   = {2012}
}

Comments

In proceedings of The 23rd International Conference on Algorithmic Learning Theory (ALT 2012)

R2 v1 2026-06-21T22:03:59.078Z