arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.07437cs.GTcs.AIcs.LGcs.MA
收藏

非原子拥塞博弈中最优税收设计的学习

Learning Optimal Tax Design in Nonatomic Congestion Games

University of Washington(华盛顿大学)

Qiwen Cui, Maryam Fazel, Simon S. Du

AI总结:

针对拥塞博弈中自利行为损害社会福利且现有算法不适用均衡反馈场景的问题,提出含分段线性税收等组件的高效算法,以低样本复杂度学习最优税收。

摘要:

在多人博弈中,参与者的自利行为会损害社会福利。税收机制是缓解这一问题、引导社会最优行为的常用方法。本研究迈出了在拥塞博弈中以有限反馈学习可最大化社会福利的最优税收的第一步。我们提出了一种名为“均衡反馈”的新型反馈机制,其中税收设计者仅能在部署税收方案后观测到纳什均衡。由于税收函数空间呈指数级规模、梯度不存在以及目标函数非凸,现有算法并不适用。为应对这些挑战,我们设计了一种计算高效的算法,它依托若干新颖组件:(1)用分段线性税收逼近最优税收;(2)引入额外线性项以保证势函数强凸;(3)一种高效子程序,用于寻找能提供博弈关键信息的探索性税收。该算法能以$O(βF^2/ε)$的样本复杂度找到$ε$-最优税收,其中$β$为成本函数的光滑度,$F$为设施数量。

英文摘要:

In multiplayer games, self-interested behavior among the players can harm the social welfare. Tax mechanisms are a common method to alleviate this issue and induce socially optimal behavior. In this work, we take the initial step of learning the optimal tax that can maximize social welfare with limited feedback in congestion games. We propose a new type of feedback named \emph{equilibrium feedback}, where the tax designer can only observe the Nash equilibrium after deploying a tax plan. Existing algorithms are not applicable due to the exponentially large tax function space, nonexistence of the gradient, and nonconvexity of the objective. To tackle these challenges, we design a computationally efficient algorithm that leverages several novel components: (1) a piece-wise linear tax to approximate the optimal tax; (2) extra linear terms to guarantee a strongly convex potential function; (3) an efficient subroutine to find the exploratory tax that can provide critical information about the game. The algorithm can find an $ε$-optimal tax with $O(βF^2/ε)$ sample complexity, where $β$ is the smoothness of the cost function and $F$ is the number of facilities.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑