Asymptotically Optimal Strategies For Combinatorial Semi-Bandits in Polynomial Time
Machine Learning
2021-02-16 v1 Machine Learning
Abstract
We consider combinatorial semi-bandits with uncorrelated Gaussian rewards. In this article, we propose the first method, to the best of our knowledge, that enables to compute the solution of the Graves-Lai optimization problem in polynomial time for many combinatorial structures of interest. In turn, this immediately yields the first known approach to implement asymptotically optimal algorithms in polynomial time for combinatorial semi-bandits.
Keywords
Cite
@article{arxiv.2102.07254,
title = {Asymptotically Optimal Strategies For Combinatorial Semi-Bandits in Polynomial Time},
author = {Thibaut Cuvelier and Richard Combes and Eric Gourdin},
journal= {arXiv preprint arXiv:2102.07254},
year = {2021}
}
Comments
26 pages