Title :
Evolutionary intelligent agents
Author :
Cristea, Paul ; Arsene, Aurel ; Nitulescu, Bogdan
Author_Institution :
Dept. of Eng. Sci., Bucharest Univ., Romania
Abstract :
Evolutionary intelligent agents (EIA) are intelligent agents provided with a genotype that controls their capabilities to carry out various tasks, i.e., their phenotype. EIA are a cognitive and genetically controlled version of ants and bees, with links to both artificial intelligence and neural networks. The evolution of the population is task dependent and favors the predominance of the individuals best fitted for the problem at hand, while preserving the genetic variety. The paper presents preliminary results in exploring the EIA concept. A distributed Java software package that implements the concept is described
Keywords :
Java; evolutionary computation; neural nets; software agents; software packages; artificial intelligence; distributed Java software package; evolutionary intelligent agents; genotype; neural networks; phenotype; Artificial intelligence; Artificial neural networks; Computer science; Distributed computing; Genetics; Intelligent agent; Java; Robustness; Software packages; Software systems;
Conference_Titel :
Evolutionary Computation, 2000. Proceedings of the 2000 Congress on
Conference_Location :
La Jolla, CA
Print_ISBN :
0-7803-6375-2
DOI :
10.1109/CEC.2000.870804
