Künstliche Intelligenz
1/06




KI Konferenz



KI Fachbereich

Böttcher IT Verlag

Heft 1/06

Schwerpunkt: Best Papers 2005

Emergent Roles in Multi Agent Systems - A Case Study in Minority Games

Zusammenfassung:

Binary decisions are common in our daily lives and often individuals can gain by choosing the minority’s side. The socio–economically inspired Minority Game (MG) has been introduced as an exact model of the famous El Farol’s Bar Problem, which exhibits complex behavior. In this paper we introduce a novel stochastic, adaptive MG. Optimal system behavior as well as individual fair behavior evolves by self–organization spawning a role called „Schwarzer Peter“. Since this role is solely caused by the system dynamics, we name it an emergent role. A challenge for Agent Oriented Software Engineering (AOSE) is the purposeful construction of these emergent roles in order to controll the behavior in Multi–Agent Systems (MAS). Therefore we present a mesoscopic modeling approach, in order to comprehend and implement similar system behavior. We show how this intermediate description level can be used to redesign the emergent role in terms of a deliberative agent implementation with equivalent system dynamics.

Autoren:

Seite/n:

25-31

Nummer:

1

Back

 

Email: info(a)kuenstliche-intelligenz.de

©KI