UCB-SC: A Fast Variant of KL-UCB-SC for Budgeted Multi-Armed Bandit Problem

Ryo WATANABE  Junpei KOMIYAMA  Atsuyoshi NAKAMURA  Mineichi KUDO  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E101-A   No.3   pp.662-667
Publication Date: 2018/03/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E101.A.662
Type of Manuscript: LETTER
Category: Mathematical Systems Science
budgeted multi-armed bandits,  regret analysis,  upper confidence bound,  

Full Text: PDF(967.8KB)>>
Buy this Article

We propose a policy UCB-SC for budgeted multi-armed bandits. The policy is a variant of recently proposed KL-UCB-SC. Unlike KL-UCB-SC, which is computationally prohibitive, UCB-SC runs very fast while keeping KL-UCB-SC's asymptotical optimality when reward and cost distributions are Bernoulli with means around 0.5, which are verified both theoretically and empirically.