In quantitative decision analysis, an analyst applies mathematical models to make decisions. Frequently these models involve an optimization problem to determine the values of the decision variables, a system </spa...In quantitative decision analysis, an analyst applies mathematical models to make decisions. Frequently these models involve an optimization problem to determine the values of the decision variables, a system </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> of possibly non</span></span><span style="font-family:Verdana;">- </span><span style="font-family:Verdana;">li</span><span style="font-family:Verdana;">near inequalities and equalities to restrict these variables, or both. In this</span><span style="font-family:""><span style="font-family:Verdana;"> note, </span><span style="font-family:Verdana;">we relate a general nonlinear programming problem to such a system </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> in</span><span style="font-family:Verdana;"> such </span><span style="font-family:Verdana;">a way as to provide a solution of either by solving the other—with certain l</span><span style="font-family:Verdana;">imitations. We first start with </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> and generalize phase 1 of the two-phase simplex method to either solve </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> or establish that a solution does not exist. A conclusion is reached by trying to solve </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> by minimizing a sum of artificial variables subject to the system </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> as constraints. Using examples, we illustrate </span><span style="font-family:Verdana;">how this approach can give the core of a cooperative game and an equili</span><span style="font-family:Verdana;">brium for a noncooperative game, as well as solve both linear and nonlinear goal programming problems. Similarly, we start with a general nonlinear programming problem and present an algorithm to solve it as a series of systems </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> by generalizing the </span></span><span style="font-family:Verdana;">“</span><span style="font-family:Verdana;">sliding objective</span><span style="font-family:Verdana;"> function </span><span style="font-family:Verdana;">method</span><span style="font-family:Verdana;">”</span><span style="font-family:Verdana;"> for</span><span style="font-family:Verdana;"> two-dimensional linear programming. An example is presented to illustrate the geometrical nature of this approach.展开更多
为解决当前竞技二打一比赛中存在的赛制冗长、存在作弊隐患的问题,设计一种牌力评估与同等牌力生成系统。对大量人类打牌数据使用二阶聚类算法进行聚类分析,得到手牌牌力分类的标注数据集;构建基于注意力机制的长短期记忆网络(long shor...为解决当前竞技二打一比赛中存在的赛制冗长、存在作弊隐患的问题,设计一种牌力评估与同等牌力生成系统。对大量人类打牌数据使用二阶聚类算法进行聚类分析,得到手牌牌力分类的标注数据集;构建基于注意力机制的长短期记忆网络(long short term memory network,LSTM)模型,针对手牌序列与手牌牌力进行训练,并通过生成同等牌力手牌借助AI机器人进行对打实验。实验结果表明:生成的不同牌力手牌在胜率上具有显著区别,可正确生成对应牌力手牌,具有可行性。展开更多
文摘In quantitative decision analysis, an analyst applies mathematical models to make decisions. Frequently these models involve an optimization problem to determine the values of the decision variables, a system </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> of possibly non</span></span><span style="font-family:Verdana;">- </span><span style="font-family:Verdana;">li</span><span style="font-family:Verdana;">near inequalities and equalities to restrict these variables, or both. In this</span><span style="font-family:""><span style="font-family:Verdana;"> note, </span><span style="font-family:Verdana;">we relate a general nonlinear programming problem to such a system </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> in</span><span style="font-family:Verdana;"> such </span><span style="font-family:Verdana;">a way as to provide a solution of either by solving the other—with certain l</span><span style="font-family:Verdana;">imitations. We first start with </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> and generalize phase 1 of the two-phase simplex method to either solve </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> or establish that a solution does not exist. A conclusion is reached by trying to solve </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> by minimizing a sum of artificial variables subject to the system </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> as constraints. Using examples, we illustrate </span><span style="font-family:Verdana;">how this approach can give the core of a cooperative game and an equili</span><span style="font-family:Verdana;">brium for a noncooperative game, as well as solve both linear and nonlinear goal programming problems. Similarly, we start with a general nonlinear programming problem and present an algorithm to solve it as a series of systems </span><i><span style="font-family:Verdana;">S</span></i><span style="font-family:Verdana;"> by generalizing the </span></span><span style="font-family:Verdana;">“</span><span style="font-family:Verdana;">sliding objective</span><span style="font-family:Verdana;"> function </span><span style="font-family:Verdana;">method</span><span style="font-family:Verdana;">”</span><span style="font-family:Verdana;"> for</span><span style="font-family:Verdana;"> two-dimensional linear programming. An example is presented to illustrate the geometrical nature of this approach.
文摘为解决当前竞技二打一比赛中存在的赛制冗长、存在作弊隐患的问题,设计一种牌力评估与同等牌力生成系统。对大量人类打牌数据使用二阶聚类算法进行聚类分析,得到手牌牌力分类的标注数据集;构建基于注意力机制的长短期记忆网络(long short term memory network,LSTM)模型,针对手牌序列与手牌牌力进行训练,并通过生成同等牌力手牌借助AI机器人进行对打实验。实验结果表明:生成的不同牌力手牌在胜率上具有显著区别,可正确生成对应牌力手牌,具有可行性。