Multi‐Armed Bandits, Gittins Index, and its Calculation
Multi‐Armed Bandits, Gittins Index, and its Calculation
复制标题
多臂老虎机、Gittins 指数及其计算
DOI：
10.1002/9781118596333.ch24
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
Aditya Mahajan
中科院分区：
文献类型：
--
作者：
Jhelum Chakravorty;Aditya Mahajan
多臂强盗是一个丰富多彩的术语，它是指在赌场里玩的赌徒面对的di引理，该赌场带有多台老虎机（通俗地称为肉食匪徒）。赌徒是否应该使用策略来挑选机器接下来玩？这是赢得后部平均值最高的均值，从而最大程度地提高了当前的预期奖励，或者是获胜后差异最高的一种，并且有可能使未来的预期奖励最大化。在各种应用领域中出现了Simi Lar Exploration vs剥削权衡取舍，包括Cl Ilicals [5]，随机调度[25]，传感器管理[33]和经济学[4]。
Multi-armed bandit is a colorful term that refers to the di lemma faced by a gambler playing in a casino with multiple slot machines (which were colloquially called onearmed bandits). W h a t strategy should a gambler use to pick the machine to play next? It is the one for which the posterior mean of winning is the highest and thereby maximizes current expected reward, or the one for which the posterior variance of winning is the highest, and has the potential to maximize the future expected reward. S imi lar exploration vs exploitation trade-offs arise in various appli cation domains including cl inical tr ials [5], stochastic scheduling [25], sensor management [33], and economics [4].