A Reinforcement Learning Approach in Multi-Phase Second-Price Auction Design
多阶段第二价格拍卖设计中的强化学习方法
机构 * Institute for Data, Systems, and Society(数据、系统与社会研究所) ; Massachusetts Institute of Technology(麻省理工学院) ; Booth School of Business(博世商学院) ; The University of Chicago(芝加哥大学) ; Department of Industrial Engineering and Management Sciences(工业工程与管理科学系) ; Northwestern University(西北大学) ; Department of Statistics and Data Science(统计与数据科学系) ; Yale University(耶鲁大学) ; Department of EECS, Department of Statistics(电子工程与计算机科学系、统计学系) ; University of California Berkeley(加州大学伯克利分校)
AI总结 本文提出CLUB算法,通过缓冲期和LSVI-UCB扩展,解决多阶段拍卖中保留价优化的三个挑战,实现不同噪声条件下收入遗憾的最优控制。