论文部分内容阅读
多智能体模型的构建可分为若干环节,其中状态更新规则往往被设定为同步状态更新规则。然而,现实自然和社会系统中普遍存在的是异步状态更新规则。已有文献发现异步状态更新规则下和同步状态更新规则下的系统动力学过程有着显著区别,但目前尚实证或实验研究探索异步性的性质。为此,基于Pearson相关系数定义了状态更新序列间一致性度量指标,然后基于由随机异步状态更新规则产生的序列间一致性频率分布,构建了判定状态更新规则有序性的方法。最后,将所构建方法应用于一组二阶段任务更新实验数据,探索性分析了现实系统的异步状态更新规则性质。结果发现,大部分系统的状态更新规则具有随机有序性。研究结论可帮助理解社会系统动力学过程,同时对多智能体建模具有重要启示。
The construction of multi-agent model can be divided into several aspects, of which the state update rule is often set as the synchronization state update rule. However, ubiquitous in the real natural and social systems is the rule of asynchronous state updating. It has been found in the literature that there is a significant difference between the system dynamics under the asynchronous state update rule and the synchronous state update rule. However, there is still empirical or experimental research to explore the nature of the asynchronous. To this end, we define the consistency metric between state update sequences based on Pearson correlation coefficient, and then construct a method to judge the orderliness of state update rules based on the frequency distribution of consistent sequences generated by random asynchronism state update rules. Finally, the constructed method is applied to a set of two-phase task updating experimental data to explore the properties of the real-time asynchronous state updating rules of real systems. The results show that most of the system’s state update rules are random and orderly. The conclusions of the study can help to understand the social system dynamics process, at the same time, it has important implications for multi-agent modeling.