【作者】 张海峰； 杨涵新； Michael Small； 汪秉宏；

【Author】 Zhang Haifeng1,2,Yang Hanxin1,Michael Small2,Wang Binghong1 1University of Science and Technology of China,Hefei Anhui,230026,China 2Hong Kong Polytechnic University,Hung Hom,Kowloon,Hong Kong

【机构】 中国科学技术大学； 香港理工大学；

【摘要】 现实中人们并不是一味的向别人学习,每个个体向别人学习的能动性总是与自身的期望收益有关,如果个体对收益的期望值很高则他向别人学习的动机更强,反之更弱。鉴于以上考虑,我们在囚徒困境博弈中引入不均匀学习动机,且这种学习动机是随着时间不断变化的,因为个体的收益是随着时间变化的。通过引入这种更加合理的机制,我们发现合作频率可以在很大范围内得到提高,并且出现非单调现象:太大的期望和太小的期望都不利于合作。最后我们研究噪音对合作频率的影响,发现存在最优的噪音使得合作频率得到提高。我们的研究结果对于理解自然界中广泛的合作现象有一定的指导意义。更多还原

【Abstract】 Evolutionary cooperation in prisoner’s dilemma game is studied where each player can adjust its learning motivation according to its actual payoff and payoff aspiration at each time step,the higher payoff aspiration he/she wants,the higher learning motivation he/she has,and vice verse.Such mechanism not only can well character the diversity of different individuals but also can promote the frequency of cooperator in spatial prisoner’s dilemma game.Additionally,the influence of noise is also studied,and we show that there exists an optimal value for noise which can promote cooperation better.Our work may shed some new light on the study of evolutionary.更多还原