01283nas a2200169 4500008004100000245005900041210005800100260000900158300001600167490000700183520079000190100001800980700001600998700001701014700002301031856005901054 2021 eng d00aQuantum exploration algorithms for multi-armed bandits0 aQuantum exploration algorithms for multiarmed bandits c2021 a10102-101100 v353 a
Identifying the best arm of a multi-armed bandit is a central problem in bandit optimization. We study a quantum computational version of this problem with coherent oracle access to states encoding the reward probabilities of each arm as quantum amplitudes. Specifically, we show that we can find the best arm with fixed confidence using O~(∑ni=2Δ−2i−−−−−−−−√) quantum queries, where Δi represents the difference between the mean reward of the best arm and the ith-best arm. This algorithm, based on variable-time amplitude amplification and estimation, gives a quadratic speedup compared to the best possible classical result. We also prove a matching quantum lower bound (up to poly-logarithmic factors).
1 aWang, Daochen1 aYou, Xuchen1 aLi, Tongyang1 aChilds, Andrew, M. uhttps://ojs.aaai.org/index.php/AAAI/article/view/17212