Attraction and cooperation in space

Tzafestas, ES

dc.contributor.author	Tzafestas, ES	en
dc.date.accessioned	2014-03-01T02:51:03Z
dc.date.available	2014-03-01T02:51:03Z
dc.date.issued	2007	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/35322
dc.subject.other	Attraction mechanism	en
dc.subject.other	Final agent's behavior	en
dc.subject.other	Game theory	en
dc.subject.other	Mathematical models	en
dc.subject.other	Parameter estimation	en
dc.subject.other	Social aspects	en
dc.subject.other	Interactive computer graphics	en
dc.title	Attraction and cooperation in space	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/CEC.2007.4424952	en
heal.identifier.secondary	http://dx.doi.org/10.1109/CEC.2007.4424952	en
heal.identifier.secondary	4424952	en
heal.publicationDate	2007	en
heal.abstract	This paper presents an attraction mechanism and a study of its behavior in noisy spatial IPD games. The attraction mechanism is coupled with a regular IPD strategy to produce the final agent's behavior. We study the mechanism in 2D space to understand how it influences the emerging spatial structures. We find that any agent strategy, even a plainly irrational one, may become stable in a spatial game given appropriate attraction relations with the neighbors. Also, often the removal of attraction after stabilization causes no qualitative effect. Then we study the effect of introduction of ""extreme"" agents, such as a ""Don Juan"" agent that is attracted by all others. Such agents can change dramatically the structure of emerging spatial strategy blocks. Especially the addition or removal of agents after stabilization almost always leads to re-organization and re-stabilization to a new configuration. Various parameters of this model are studied and finally the notion of mobility in relation to evolution is rediscussed. All our results suggest that psychological mechanisms external to the actual ""problem"" (here IPD) can interfere with it and can actually lead to enhanced cooperation and social stability despite environmental noise and agent irrationality. Moreover, all social/spatial variants that induce more interactions between agents with non-reciprocal attraction relations (such as the extreme agents), are bound to lead to a better social average and to more complex structures. © 2007 IEEE.	en
heal.journalName	2007 IEEE Congress on Evolutionary Computation, CEC 2007	en
dc.identifier.doi	10.1109/CEC.2007.4424952	en
dc.identifier.spage	3698	en
dc.identifier.epage	3705	en