Influence of network topology on cooperative problem-solving systems
- 337 Downloads
The idea of a collective intelligence behind the complex natural structures built by organisms suggests that the organization of social networks is selected so as to optimize problem-solving competence at the group level. Here we study the influence of the social network topology on the performance of a group of agents whose task is to locate the global maxima of NK fitness landscapes. Agents cooperate by broadcasting messages informing on their fitness and use this information to imitate the fittest agent in their influence networks. In the case those messages convey accurate information on the proximity of the solution (i.e., for smooth fitness landscapes), we find that high connectivity as well as centralization boosts the group performance. For rugged landscapes, however, these characteristics are beneficial for small groups only. For large groups, it is advantageous to slow down the information transmission through the network to avoid local maximum traps. Long-range links and modularity have marginal effects on the performance of the group, except for a very narrow region of the model parameters.
KeywordsImitative learning Group size Social networks
The research of JFF was partially supported by grant 2013/17131-0, São Paulo Research Foundation (FAPESP), and by grant 303979/2013-5, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). FAR acknowledges CNPq (grant 305940/2010-4) and FAPESP (grant 2013/26416-9) for financial support.
- Axelrod R (1984) The evolution of cooperation. Basic Books, New YorkGoogle Scholar
- Bloom H (2001) Global brain: the evolution of mass mind from the big bang to the 21st century. Wiley, New YorkGoogle Scholar
- Bonabeau E, Dorigo M, Theraulaz G (1999) Swarm Intelligence: from natural to artificial systems. Oxford University Press, OxfordGoogle Scholar
- Englemore R, Morgan T (1988) Blackboard Sys. Addison- Wesley, New YorkGoogle Scholar
- Garey MR, Johnson DS (1979) Computers and intractability: a guide to the theory of NP-completeness. Freeman, San FranciscoGoogle Scholar
- Moffett MW (2011) Adventures among ants: a global safari with a cast of trillions. University of California Press, OaklandGoogle Scholar
- Pasquaretta C, Levé M, Claidière N, van de Waal E, Whiten A, MacIntosh AJJ, Pelé M, Bergstrom ML, Borgeaud C, Brosnan SF, Crofoot MC, Fedigan LM, Fichtel C, Hopper LM, Mareno MC, Petit O, Schnoell AV, di Sorrentino EP, Thierry B, Tiddi B, Sueur C (2014) Social networks in primates: smart and tolerant species have more efficient networks. Sci Rep 4:7600CrossRefPubMedPubMedCentralGoogle Scholar
- Solow D, Burnetas A, Tsai M, Greenspan NS (2000) On the expected performance of systems with complex interactions among components. Complex Sys 12:423–456Google Scholar
- Wilson E (1975) Sociobiology. Harvard University Press, CambridgeGoogle Scholar