为了研究在复杂社会网络环境中的社会学习,考虑到社会个体的异质性和复杂性,提出一种基于复合策略进行信念更新的社会学习模型。模型中的个体在每一时刻,以一定概率任意选择基于贝叶斯法则的更新策略或基于邻居信念的更新策略中的一种策略进行信念更新。仿真结果表明,在满足策略选择概率大于0等条件下,社会个体最终都能够达到渐近学习,并且学习速度与策略选择概率是相关的,策略选择概率取值越大,学习速度越快。
In order to study social learning in complex social networks, a social learning model based on composite belief update strategy is proposed by considering the heterogeneity and complexity of the social individuals. At each time step, individuals in the model choose Bayesian update strategy or the update strategy based on their neighbors’ beliefs according to the strategy selection probability. The simulation results show that under some conditions such as the positive strategy selection probability, all the social individuals can achieve asymptotic learning. Furthermore, the learning speed is relative to the strategy selection probability, the larger the strategy selection probability is, the faster the learning speed will be.
[1] Albert B. Social learning Theory[M]. London: Prentice Hall, 1976.
[2] Lee I. On the convergence of informational cascades[J]. The Journal of Economic Theory, 1993, 61(2): 395-411.
[3] Tsitsiklis J N, Athans M. Convergence and asymptotic agreement in distributed decision probl-ems[C]. IEEE Transactions on Automatic Control, 1984, 29(8): 690-696.
[4] Feddersen T, Pesendorfer W. Voting behavior and information aggregation in elections with private information[J]. Econometrica, 1997, 65(5): 1029-1058.
[5] Foster A, Rosenzweig M. Learning by doing and learning from others: human capital and technical change in agriculture[J]. The Journal of Political Economy, 1995, 103(6): 1176-1209.
[6] Celen B, Kariv S. Observational learning under imperfect information[J].Games and Economic Behavior, 2004, 47(1): 72-86.
[7] Borkar V, Varaiya P. Asymptotic agreement in distributed estimation[C]. IEEE Transactions on Automatic Control, 1982, 27: 650-655.
[8] Udry C, Conley T. Social learning through networks: the adoption of new agricultural technologies in ghana[J]. American Journal of Agricultural Economics, 2001, 83(3): 668-673.
[9] Banerjee A. A simple model of herd behavior[J]. The Quarterly Journal of Economics, 1992, 107(3): 797-817.
[10] Smith L, Sørensen P. Pathological outcomes of observational learning[J]. Econometrica, 2000, 68(2): 371-398.
[11] Gale D, Kariv S. Bayesian learning in social networks[J].Games and Economic Behavior, 2003, 45(2): 329-346.
[12] Banerjee A, Fudenberg D. Word-of-mouth learning[J]. Games and Economic Behavior, 2004, 46(1):1-22
[13] Celen B, Kariv S. An experimental test of observational learning under imperfect information[J]. Economic Theory, 2005,26(3): 677-699.
[14] Montrey M R, Shultz T R. Evolution of social learning strategies[C]. 2010 IEEE 9th International Conference on Development and learning, Ann Arbor, MI, USA, 2010: 95-100.
[15] DeGroot M H. Reaching a consensus[J]. Journal of the American Statistical Association, 1974, 69(345): 118-121.
[16] Ellision Glenn, Fudenberg D. Rules of thumb for social iearning[J]. The Journal of Political Economy, 1993, 101(4): 612-643.
[17] Ellision Glenn, Fudenberg D. Word-of-mouth communication and social learning[J]. Quarterly Journal of Economics, 1995, 110(1): 93-125.
[18] Bala V, Goyal S. Learning from neighbors[J]. Review of Economic Studies, 1998, 65(3):595-621.
[19] Bala V, Goyal S. Conformism and diversity under social learning[J]. Economic Theory, 2001, 17(1): 101-120.
[20] Golub B, Jackson M. Naive learning in social networks and the wisdom of crowds[J]. American Economic Journal: Microeconomics, 2010, 2(1): 112-149.
[21] Jadbabaie A, Molavi P, Sandroni A, et al. Non-Bayesian social learning[J]. Games and Economic Behavior, 2012, 76(1): 210-225.
