Optimal Learning for Structured Bandits: DUSA Has Asymptotic Regret at Most (1+ε)(C(P)+ε(1+Σₓ∆(x,P))) log T, Matching the Lower Bound C(P) as ε → 0, with O(log T) Exploration RoundsResearch Paper
In a multi-armed bandit a learner repeatedly pulls one of finitely many arms and observes a random reward, trying to earn as much as an omniscient learner who knows the best arm. In many operations problems the arms are not unrelated: in dynamic pricing the probability of no sale is monotone in the price, in assortment or dosage problems mean rewards vary…