arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2312.02668cs.GTmath.DS
收藏

Shapley网络设计博弈的近似纳什均衡算法

Approximate Nash Equilibria Algorithms for Shapley Network Design Games

Hangxin Gan, Xianhao Meng, Chunying Ren, Yongtang Shi

AI总结:

针对加权Shapley网络设计博弈的纯纳什均衡求解问题,证明了α-近似均衡的存在性并给出α的上界,还证明了φ-光滑随机成本下ε-最优响应动力学可多项式时间计算(1+ε)α-近似均衡。

摘要:

我们研究加权Shapley网络设计博弈,其中自私的参与者通过选择网络中的路径来最小化自身成本。网络中每条边的成本函数相对于选择该边的参与者的权重之和是仿射线性的。我们首先通过构造势函数证明了α-近似纯纳什均衡的存在性,并建立了α的上界O(log₂(W)),其中W为所有参与者的权重之和。此外,我们假设边的成本函数(仿射线性函数)的系数均为[0,1]上的φ-光滑随机变量。在该设定下,我们通过证明迭代次数的期望是1/ε、φ、参与者数量和网络边数的多项式,表明ε-最优响应动力学可在多项式时间内计算出(1+ε)α-近似纯纳什均衡(ε为接近0的正常数)。

英文摘要:

We consider a weighted Shapley network design game, where selfish players choose paths in a network to minimize their cost. The cost function of each edge in the network is affine linear with respect to the sum of weights of the players who choose the edge. We first show the existence of α-approximate pure Nash equilibrium by constructing a potential function and establish an upper bound O(log2(W)) of α, where W is the sum of the weight of all players. Furthermore, we assume that the coefficients of the cost function (affine linear function) of the edge all are ϕ-smooth random variables on [0, 1]. In this case, we show that ε-best response dynamics can compute the (1 + ε)α-approximate pure Nash equilibrium (εis a positive constant close to 0) in polynomial time by proving the expected number of iterations is polynomial in 1/ε, ϕ, the number of players and the number of edges in the network.

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。

↑