How to Beat the Adaptive Multi-Armed Bandit

The multi-armed bandit is a concise model for the problem of iterated decision-making under uncertainty. In each round, a gambler must pull one of K arms of a slot machine, without any foreknowledge of their payouts, except that they are uniformly bounded. A standard objective is to minimize the gambler’s regret, de ned as the gambler’s total payout minus the largest payout which would have been achieved by any xed arm, in hindsight. Note that the gambler is only told the payout for the arm actually chosen, not for the unchosen arms. Almost all previous work on this problem assumed the payouts to be non-adaptive, in the sense that the distribution of the payout of arm j in round i is completely independent of the choices made by the gambler on rounds 1; : : : ; i 1. In the more general model of adaptive payouts, the payouts in round i may depend arbitrarily on the history of past choices made by the algorithm. We present a new algorithm for this problem, and prove nearly optimal guarantees for the regret against both non-adaptive and adaptive adversaries. After T rounds, our algorithm has regret O( p T ) with high probability (the tail probability decays exponentially). This dependence on T is best possible, and matches that of the full-information version of the problem, in which the gambler is told the payouts for all K arms after each round. Previously, even for non-adaptive payouts, the best high-probability bounds known were O(T 2=3), due to Auer, Cesa-Bianchi, Freund and Schapire [1]. For non-adaptive payouts, they also proved an O( p T ) bound on expected regret. We describe an adaptive payout scheme for which the expected regret of their algorithm is (T 2=3).