Title :
A comparison of team evolution operators
Author :
Qi, Dehu ; Sun, Ron
Author_Institution :
Dept. of Comput. Sci., Lamar Univ., Beaumont, TX, USA
Abstract :
One of the main research topics in multi-agent systems is learning cooperation among agents. In the MARLBS system, we use genetic algorithms to evolve neural networks, which enhances the cooperation between agents. In this paper, we examine several evolutionary operators to evolve a team, in which team members cooperate with each other to solve problems. The best operators found from experiments efficiently reduce learning time.
Keywords :
genetic algorithms; learning (artificial intelligence); multi-agent systems; neural nets; MARLBS system; agents cooperation; evolutionary operators; genetic algorithms; learning cooperation; multi-agent systems; neural networks; team evolution operators; Cognitive science; Computer science; Costs; Evolutionary computation; Genetic algorithms; Learning; Multiagent systems; Neural networks; Neurons; Sun;
Conference_Titel :
Intelligent Agent Technology, 2004. (IAT 2004). Proceedings. IEEE/WIC/ACM International Conference on
Print_ISBN :
0-7695-2101-0
DOI :
10.1109/IAT.2004.1342973
