Fetching the paper…

UCB algorithms for multi-armed bandits: Precise regret and adaptive inference · Around