We investigate performance of the Genetic Landscape Evolution (GLE) model by changing number of crossover points, which controls spatial cohesiveness of topological information in generated offspring. Simulation results show that 1) GLE performance is insensitive to the number of crossover points, implying that the spatial cohesiveness does not significantly affect efficiency to find better solution sets; and 2) the method to generate randomness in GLE is a significant element for its performance.
CITATION STYLE
Kim, J., & Paik, K. (2016). Performance evaluation of the Genetic Landscape Evolution (GLE) model with respect to crossover schemes. In Advances in Intelligent Systems and Computing (Vol. 382, pp. 377–383). Springer Verlag. https://doi.org/10.1007/978-3-662-47926-1_36
Mendeley helps you to discover research relevant for your work.