Unified theory of upper confidence bound policies for bandit problems targeting total reward, maximal reward, and more

Re-assign community