We propose an algorithm for computing Nash equilibria (NE) in a class of conflicts with multiple battlefields with uniform battlefield values and a non-linear aggregation function. By expanding the symmetrization idea of Hart [9], proposed for the Colonel Blotto game, to the wider class of symmetric conflicts with multiple battlefields, we reduce the number of strategies of the players by an exponential factor. We propose a clash matrix algorithm which allows for computing the payoffs in the symmetrized model in polynomial time. Combining symmetrization and clash matrix algorithm with the double oracle algorithm we obtain an algorithm for computing NE in the models in question that achieves a significant speed-up as compared to the standard, LP-based, approach. We also introduce a heuristic to further speed up the process. Overall, our approach offers an efficient and novel method for computing NE in a specific class of conflicts, with potential practical applications in various fields.
Johannes Kepler University Linz(林茨约翰·开普勒大学);Austrian Institute of Economic Research(奥地利经济研究所);Inria/FairPlay(法国国家信息与自动化研究所/公平游戏实验室);Tel Aviv University(特拉维夫大学);Maison Française d’Oxford, CNRS and Department of Economics, University of Oxford(牛津法国之家、法国国家科学研究中心及牛津大学经济学系)
A Game-Theoretic Analysis of the Empirical Revenue Maximization Algorithm with Endogenous Sampling
Peking University(北京大学);Technion - Israel Institute of Technology(以色列理工学院);Hong Kong University of Science and Technology(香港科技大学);Alibaba Group(阿里巴巴集团);Shanghai Jiao Tong University(上海交通大学)
经验收益最大化(ERM)是拍卖设计中最重要的价格学习算法之一:文献表明,它可以在重复拍卖和统一价格拍卖中学习到收益最大化拍卖者的近似最优保留价。然而,在这些应用程序中,提供输入的机构有动机操纵输入,以降低输出的价格。我们概括了Lavi et al(2019)提出的激励意识度量的定义,以量化由于$N$输入样本中的$m\ge 1$的变化而导致的ERM输出价格的降低,并提供了该度量到零的特定收敛率,因为对于不同类型的输入分布,$N$趋于无穷大。通过采用这一措施,我们构建了一个有效的,近似激励兼容的,和收入最优的学习算法使用ERM在重复拍卖对非近视投标人,并显示近似组激励兼容统一价格拍卖。
Learning Safe Strategies for Value Maximizing Buyers in Uniform Price Auctions
Sloan School of Management, Massachusetts Institute of Technology(麻省理工学院斯隆管理学院);Operations Research Center, Massachusetts Institute of Technology(麻省理工学院运筹学研究中心)