Title :
Minority games and distributed coordination in non-stationary environments
Author :
Galstyan, Aram ; Lerman, Kristina
Author_Institution :
Inf. Sci. Inst., Univ. of Southern California, Marina del Rey, CA, USA
fDate :
6/24/1905 12:00:00 AM
Abstract :
We examine emergent coordination in a network of competing Boolean agents. The agents play a so called generalized minority game where the capacity level is allowed to vary externally. We study the properties of such a system for different values of the mean connectivity K of the network, and show that the system with K = 2 shows a high degree of coordination for relatively large variations of the capacity level. We also show that for K > 2 coordination can be achieved by tuning the homogeneity parameter of the agents´ Boolean strategies
Keywords :
Boolean functions; adaptive systems; game theory; multi-agent systems; capacity level; competing Boolean agents; distributed coordination; emergent coordination; homogeneity parameter; minority games; nonstationary environments; Artificial intelligence; Centralized control; Defense industry; Distributed control; Intelligent agent; Intelligent networks; Intelligent robots; Intelligent sensors; Multiagent systems; Robustness;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007555