Title :
Social learning in a society of decentralized agents
Author_Institution :
Dept. of Comput. Sci., Nat. Defense Acad., Kanagawa, Japan
Abstract :
The paper describes research into social learning leading to the coordinated behavior of decentralized agents. We term such an agent, having both a selfish interest and social competence, a decentralized agent. We obtain equilibrium solutions under both individually rational and socially rational behavior of decentralized agents. We formulate the model of social learning as a mutual adjustment process of decision making in a society of decentralized agents. We discuss how the coordinated and socially optimal behavior can emerge from mutual interactions based on the selfish interest seeking of decentralized agents
Keywords :
cooperative systems; decision theory; knowledge based systems; learning (artificial intelligence); software agents; coordinated behavior; decentralized agent society; decision making; equilibrium solutions; individually rational behavior; mutual adjustment process; mutual interactions; selfish interest; social competence; social learning; socially optimal behavior; socially rational behavior; Collaboration; Computer science; Costs; Decision making; History;
Conference_Titel :
Evolutionary Computation, 1996., Proceedings of IEEE International Conference on
Conference_Location :
Nagoya
Print_ISBN :
0-7803-2902-3
DOI :
10.1109/ICEC.1996.542387
