勒索
结果(博弈论)
成对比较
进化动力学
囚徒困境
进化博弈论
困境
计算机科学
微观经济学
经济
博弈论
进化生物学
人工智能
数学
生物
地理
社会学
人口
考古
几何学
人口学
作者
Attila Szolnoki,Matjaž Perc
出处
期刊:Physical Review E
[American Physical Society]
日期:2014-02-14
卷期号:89 (2): 022804-022804
被引量:160
标识
DOI:10.1103/physreve.89.022804
摘要
Extortion strategies can dominate any opponent in an iterated prisoner's dilemma game. But if players are able to adopt the strategies performing better, extortion becomes widespread and evolutionary unstable. It may sometimes act as a catalyst for the evolution of cooperation, and it can also emerge in interactions between two populations, yet it is not the evolutionarily stable outcome. Here we revisit these results in the realm of spatial games. We find that pairwise imitation and birth-death dynamics return known evolutionary outcomes. Myopic best response strategy updating, on the other hand, reveals counterintuitive solutions. Defectors and extortioners coarsen spontaneously, which allows cooperators to prevail even at prohibitively high temptations to defect. Here extortion strategies play the role of a Trojan horse. They may emerge among defectors by chance, and once they do, cooperators become viable as well. These results are independent of the interaction topology, and they highlight the importance of coarsening, checkerboard ordering, and best response updating in evolutionary games.
科研通智能强力驱动
Strongly Powered by AbleSci AI