English

A penalized bandit algorithm

Probability 2016-08-16 v1

Abstract

We study a two armed-bandit algorithm with penalty. We show the convergence of the algorithm and establish the rate of convergence. For some choices of the parameters, we obtain a central limit theorem in which the limit distribution is characterized as the unique stationary distribution of a discontinuous Markov process.

Keywords

Cite

@article{arxiv.math/0510384,
  title  = {A penalized bandit algorithm},
  author = {Damien Lamberton and Gilles Pagès},
  journal= {arXiv preprint arXiv:math/0510384},
  year   = {2016}
}

Comments

31 pages