Title :
Irreducible complexity in a genetic algorithm
Author :
Graham, Lee ; Oppacher, Franz ; Christensen, Steffen
Author_Institution :
Carleton Univ., Ottawa
Abstract :
There has been much news about the ability of evolution to evolve systems where function ceases with the removal of any of its multiple parts. In particular, a challenge has been lofted to explain how evolution can evolve such a system, given its gradualistic nature. We present an example utilizing a dynamic fitness function and parsimony pressure. Given unambiguous definitions of system, function, and part, the genetic algorithm presented here readily produces such complex systems.
Keywords :
computational complexity; genetic algorithms; learning (artificial intelligence); complex systems; dynamic fitness function; evolve systems; genetic algorithm complexity; gradualistic nature; parsimony pressure; Evolutionary computation; Genetic algorithms;
Conference_Titel :
Evolutionary Computation, 2007. CEC 2007. IEEE Congress on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-1339-3
Electronic_ISBN :
978-1-4244-1340-9
DOI :
10.1109/CEC.2007.4424951
