English

Optimally Confident UCB: Improved Regret for Finite-Armed Bandits

Machine Learning 2016-02-25 v3 Optimization and Control

Abstract

I present the first algorithm for stochastic finite-armed bandits that simultaneously enjoys order-optimal problem-dependent regret and worst-case regret. Besides the theoretical results, the new algorithm is simple, efficient and empirically superb. The approach is based on UCB, but with a carefully chosen confidence parameter that optimally balances the risk of failing confidence intervals against the cost of excessive optimism.

Keywords

Cite

@article{arxiv.1507.07880,
  title  = {Optimally Confident UCB: Improved Regret for Finite-Armed Bandits},
  author = {Tor Lattimore},
  journal= {arXiv preprint arXiv:1507.07880},
  year   = {2016}
}

Comments

26 pages