引入时间收益的空间公共品博弈在反省机制下的仿真分析  

Simulation Analysis of Spatial public Goods Games with Time-rewarding under the Self-questioning Mechanism

作　　者：冯瑞雪[1] 张宏斌[1] 王宏[2] 丁迪[1] 葛娟[1] 

机构地区：[1]苏州大学计算机科学与技术学院,江苏苏州215006 [2]中国地质大学数学与物理学院,湖北武汉430070

出　　处：《系统工程》2014年第2期117-122,共6页Systems Engineering

基　　金：国家自然基金资助项目(61070169)

摘　　要：在现实中,提前完成工作的奖励机制被经常采用,通过提前完成工作节省时间得到的贴现效应能够促进社会系统更好地合作。通过构建时间收益w-w/nC来反映这种时间的贴现效应,其中nC是一个博弈群中合作者的数量,w是一个时间收益参数。引入时间收益的公共品博弈模型是为了研究当更多的代理人愿意合作使任务提前完成而得到的时间效益在反省机制下对博弈演化的影响。演化结果表明,当w>0时,时间效益在乘法因子r的一个很宽的范围能明显的促进合作。尤其是对于较小的r,在w增长的初始阶段,存在一个"冰冻区间"使得系统的合作水平fc持续在最低值,但是当w接近一个阈值wc,合作水平开始迅速增加。最后,研究噪声对演化过程的影响。The incentive effects of finishing the work in advance are often employed in reality, likewise, the present discounted effect of the future profits by finishing the work sooner can also promote the cooperative level. The time-rewarding form vo -- w/nc was constructed to reflect the present discounted profits where nc hints the number of cooperators in a game group and zo stands for a time-rewarding parameters. The modified public goods games （PGG） with time-rewarding are generalized to study the effect of additional time benefit resulting from earlier completing of a task when more agents willing to share the work. The results in the self-questioning mechanism （SM） indicate that for w^0, the time-rewarding can apparently promote the cooperation over a wide range of the multiply factor r. Particularly for some smaller r, at the beginning of the growth in zo, there exists a ＂frozen interval＂ so that the cooperation frequency fc wanders at the lowest level, but when w closes to a threshold value we, the cooperation level fc begin to increase rapidly. Finally, the impact of noise value on the evolution is studied.

关 键 词：公共品博弈 时间效益 自我反省 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]