2000 character limit reached
Optimally Confident UCB: Improved Regret for Finite-Armed Bandits (1507.07880v3)
Published 28 Jul 2015 in cs.LG and math.OC
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.