Title :
Bottom up approach for deriving the redundancy of Structured Genetic Algorithms
Author :
Molfetas, Angelos
Author_Institution :
Western Sydney Univ., Sydney
Abstract :
This study examines how the redundancy of the structured genetic algorithm changes with the incorporation of control levels. Given that not all genes are activated in particular levels, the addition of control levels raises the number of redundant genes. The addition of control levels above one also increases the redundancy ratio, provided the number of genes in the top level remains fixed. The redundancy ratio, however, is not guaranteed to raise with each control level above one if the number of bottom level genes is held constant, instead of assuming a fixed number of top level genes. These are significant findings as there are strong indicators that redundancy may be correlated to algorithmic performance.
Keywords :
genetic algorithms; bottom level genes; redundant genes; structured genetic algorithms redundancy; Algorithm design and analysis; Artificial neural networks; Biological cells; Encoding; Equations; Genetic algorithms; Mathematical analysis; Mathematics; Neurons; Size control;
Conference_Titel :
Evolutionary Computation, 2006. CEC 2006. IEEE Congress on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-9487-9
DOI :
10.1109/CEC.2006.1688345