Diagonal crossover in genetic algorithms for numerical optimization. Evolutionary computation
Á. E. Eiben, C.H.M. vanKemenade · Centrum Wiskunde & Informatica (CWI), the national research institute for mathematics and computer science in the Netherlands · 1997
In this paper the results of a detailed investigation on a multi-parent recombination operator, diagonal crossover, arc reported.Although earlier publications have indicated the high performance of diagonal crossover on a number of problems, so far it has not been investigated whether high performance is indeed a result of using a high number of parents.Here we formulate three hypotheses to explain why GA performance increases when more parents arc used.Based on an extensive study on a test suite con-taiI1ing eight numerical optimization problems we arc able to establish that the higher number of parents is indeed one of the sources of higher performance, if and when this occurs.By the diversity of the test functions (unimodal, multimodal, quasi-random landscapes) we can also make observations on the relationship between fitness landscapes and operator performance.