一类多步矩阵对策上的计策问题

以有序树为工具,研究了可以描述连环计,诱敌深入等多步矩阵对策上的一类计策模型.在不考虑信息环境的封闭对策系统中,及局中人对每一步矩阵对策的赢得矩阵,两个局中人的策略集合以及局中人的理性等的了解都是局中人的共同知识的假定下,提出了局中人的最优计策链及将计就计等概念,研究了局中人中计和识破计策的固有概率,讨论了局中人在什么情况下最好主动用计,在什么情况下最好从动用计以及求解最优计策等问题.     

连续对策的最大熵策略密度对策解

对于正方形[0,2]×[0,2]上的连续对策,将局中人的非纯策略(概率分布函数)的导数称为这个局中人的策略密度(概率密度函数).建立了这种连续对策的最大熵理论.主要证明了当每个局中人都没有最优纯策略时,具有最大熵的最优策略密度集合的非空紧凸性,研究了最优策略密度的最大熵,给出一类带有最大熵的连续对策.     

多重因素约束下的网格检查对策问题研究

基于物品数量及每列容量等限制因素,构造局中人的可行策略集合;考虑隐藏成本,处罚规则与检查成功概率等因素,构造相应的支付函数,建立多重因素约束下的网格检查对策模型.根据矩阵对策性质,将对策论问题转化为非线性整数规划问题,利用H(o|¨)lder不等式获得实数条件下的规划问题的解,然后转化为整数解,得到特定条件下的模型的对策值及局中人的最优混合策略.最后,给出一个实例,说明上述模型的实用性及方法的有效性.     

矩阵对策的两个注记

设(x*,y*)是以A=[aij]m×n为赢得矩阵G的对策解,则当局中人1,2各自独立地使用其最优策略x*=(x*1,x*2,…,xmn),y*=(y*1,y*2,…,y*n)时,局中人1的赢得期望为对策值v*=x*Ay*T.若局中人双方使用使得方差D(x*,y*)=∑∑(aij-v*)2x*iy*j达最小的对策解(x*,y*),则其赢得靠近v*的概率达到最大.以O记使方差达到最小的对策解的集合.若O满足(x(1),y(1)),(x(2),y(2))∈O蕴涵(x(1),y(2)),(x(2),y(1))∈O,则说O是可换的.本文首先证明了若矩阵对策G有纯解,则O是可换的.然后证明了如果限定局中人1在其混合扩充策略集的一个非空紧凸子集X中选取策略,那么存在X的一个非空紧子集O(X),它是有限个非空互不相交紧凸集之并,使得只要局中人1使用O(X)中的策略,那么在最坏的情况下可以取得最好的赢得.     

连续对策的公平性和刺激性

本文用刺激性(感)来描述游戏一个零和对策的两个局中人的风险性和侥幸取胜性,游戏不同的零和对策可能有不同的刺激感,刺激性越大,对策结果的公平性越小;反之亦然,本文解决了如下问题;(1)刺激性和公平性的数学描述是什么?(2)局中人如何保证他们的一局对策的对策结果是最公平的或最有刺激感的?(3)如果两个局中人希望对策结果尽量公平或尽量有刺激感,他们最好从给定的连续对策中选择哪个?     

支付值为直觉模糊集的矩阵对策的线性规划求解方法

研究支付值为直觉模糊集的矩阵对策的求解方法.提出了支付值为直觉模糊集的矩阵对策的定义,并根据多目标优化的帕雷托最优解的概念定义了直觉模糊矩阵对策解的概念.进一步根据解的定义,证明了求此对策问题的解转化为求线性规划问题的最优解.通过一个数值实例说明了该方法的有效性和实用性.     

模糊矩阵对策

本文考虑了三种类型的模糊矩阵对策问题，提出了最优解集和对策值对价值标准的连续依赖性，对策双方最优策略的可调和性等概念，得到了一些基本结果。模糊矩阵对策；价值标准；连续依赖性；可调和性     

半马尔可夫对策

本文考虑半马尔可夫随机对策.在一定条件下,我们证明随机对策有值函数,两个局中人相对于折扣报酬都有最优策略.     

连续对策之判断下的最优策略集

本文引进连续对策上的判断块、判断准确、判断下的最优策略集等概念，得到了如下几个主要结果：1.判断下的最优策略集是一个局部凸空间的非空有界闭凸集；2.两个判断下的最优策略集相等的充要条件是这两个判断位于同一判断块中；3.若局中人判断准确，则在一次性对策下不论他使用此判断下的那一个最优策略（不论是纯的还是混合的），都可无风险地取得最优赢得。     

单调集对策及合成对策的边缘值

本文给出了单调集对策及其合成对策的边缘值，它类似于我们所熟知的TU—对策的Shapley值及文献[6]．集对策的边缘值的意义在于允许局中人共享项目．这使得不能分割的项目在局中人之间的分配成为可能．我们给出了这种边缘值的一些性质，并讨论了合成集对策的核及其子对策的核之间的关系．     

Optimal strategies for equal-sum dice games

In this paper, we consider a non-cooperative two-person zero-sum matrix game, called dice game. In an (n,σ) dice game, two players can independently choose a dice from a collection of hypothetical dice having n faces and with a total of σ eyes distributed over these faces. They independently roll their dice and the player showing the highest number of eyes wins (in case of a tie, none of the players wins). The problem at hand in this paper is the characterization of all optimal strategies for these games. More precisely, we determine the (n,σ) dice games for which optimal strategies exist and derive for these games the number of optimal strategies as well as their explicit form.     

Assignment games with stable core

We prove that the core of an assignment game (a two-sided matching game with transferable utility as introduced by Shapley and Shubik, 1972) is stable (i.e., it is the unique von Neumann-Morgenstern solution) if and only if there is a matching between the two types of players such that the corresponding entries in the underlying matrix are all row and column maximums. We identify other easily verifiable matrix properties and show their equivalence to various known sufficient conditions for core-stability. By these matrix characterizations we found that on the class of assignment games, largeness of the core, extendability and exactness of the game are all equivalent conditions, and strictly imply the stability of the core. In turn, convexity and subconvexity are equivalent, and strictly imply all aformentioned conditions. Final version: April 1, 2001     

矩阵博弈的胜利度和“僵局”

矩阵博弈又称有限对抗博弈，而对抗博弈的结果必是一方胜利，失败或双方和局，本首先给出胜利、失败和和局的数学模型，接着给出描述胜利程度的概念-胜利度并讨论了胜利度的几个简单性质，依此分析了传统矩阵博弈出现“僵局”的原因并指出排除“僵局”的方法。     

全对策的边际贡献值

全对策是定义在局中人集合的所有分划集上的一类特殊合作对策.本文在效用可转移情形下研究全对策的"值"问题.定义了全对策的边际贡献值,得出全对策的Shapley值,以及具有某些性质的值是边际贡献值,并给出两种边际贡献值的具体表达式,及其一些性质.     

Atomic weights and the combinatorial game of bipass

We define an all-small ruleset, bipass, within the framework of normal play combinatorial games. A game is played on finite strips of black and white stones. Stones of different colors are swapped provided they do not bypass one of their own kind. We find a simple surjective function from the strips to integer atomic weights (Berlekamp, Conway and Guy 1982) that measures the number of units in all-small games. This result provides explicit winning strategies for many games, and in cases where it does not, it gives narrow bounds for the canonical form game values. We find game values for some parametrized families of games, including an infinite number of strips of value ?, and we prove that the game value ?2 does not appear as a disjunctive sum of bipass. Lastly, we define the notion of atomic weight tameness, and prove that optimal misére play bipass resembles optimal normal play.     

A single-shot game of multi-period inspection

This paper deals with an inspection game of Customs and a smuggler during some days. Customs has two options of patrolling or not. The smuggler can take two strategies of shipping its cargo of contraband or not. Two players have several opportunities to take an action during a limited number of days but they may discard some of the opportunities. When the smuggling coincides with the patrol, there occurs one of three events: the capture of the smuggler by Customs, a success of the smuggling and nothing new. If the smuggler is captured or no time remains to complete the game, the game ends. There have been many studies on the inspection game so far by the multi-stage game model, where both players at a stage know players’ strategies taken at the previous stage. In this paper, we consider a two-person zero-sum single-shot game, where the game proceeds through multiple periods but both players do not know any strategies taken by their opponents on the process of the game. We apply dynamic programming to the game to exhaust all equilibrium points on a strategy space of player. We also clarify the characteristics of optimal strategies of players by some numerical examples.     

情绪影响下选时博弈模型及其最优策略

依据Quiggin的秩依期望效用理论研究经典选时博弈问题。通过引入可以刻画局中人在博弈中情绪状态的非线性决策权重函数,将RDEU有限策略博弈扩展到连续博弈,构建了RDEU选时博弈模型。基于Riccati微分方程的解法,求出博弈模型中局中人的最优策略。最后,通过数值仿真,分析了不同情绪状态对局中人博弈决策行为的影响。研究发现,情绪对混合策略意义下的局中人最优策略有着显著的影响,在乐观情绪状态下,局中人对混合策略极易产生自信和较高的信任倾向,表现出"风险爱好者"行为;在悲观情绪状态下,局中人往往对混合策略缺乏自信和信任,表现出“风险厌恶者”行为。     

Maker–Breaker domination game

We introduce the Maker–Breaker domination game, a two player game on a graph. At his turn, the first player, Dominator, selects a vertex in order to dominate the graph while the other player, Staller, forbids a vertex to Dominator in order to prevent him to reach his goal. Both players play alternately without missing their turn. This game is a particular instance of the so-called Maker–Breaker games, that is studied here in a combinatorial context. In this paper, we first prove that deciding the winner of the Maker–Breaker domination game is pspace-complete, even for bipartite graphs and split graphs. It is then showed that the problem is polynomial for cographs and trees. In particular, we define a strategy for Dominator that is derived from a variation of the dominating set problem, called the pairing dominating set problem.