Multi-Parent's Niche: n-ary Crossovers on NK-Landscapes

Using the multi-parent diagonal and scanning crossover in GAs reproduction operators obtain an adjustable arity. Hereby sexuality becomes a graded feature instead of a Boolean one. Our main objective is to relate the performance of GAs to the extent of sexuality used for reproduction on less arbitrary functions then those reported in the current literature. We investigate GA behaviour on Kauffman's NK-landscapes that allow for systematic characterization and user control of ruggedness of the fitness landscape. We test GAs with a varying extent of sexuality, ranging from asexual to ’very sexual’. Our tests were performed on two types of NK-landscapes: landscapes with random and landscapes with nearest neighbour epistasis. For both landscape types we selected landscapes from a range of ruggednesses. The results confirm the superiority of (very) sexual recombination on mildly epistatic problems.

[1]  Stuart A. Kauffman,et al.  ORIGINS OF ORDER IN EVOLUTION: SELF-ORGANIZATION AND SELECTION , 1992 .

[2]  Hans-Georg Beyer,et al.  Toward a Theory of Evolution Strategies: On the Benefits of Sex the (/, ) Theory , 1995, Evolutionary Computation.

[3]  Hans-Paul Schwefel,et al.  Evolution and optimum seeking , 1995, Sixth-generation computer technology series.

[4]  L. C. Stayton,et al.  On the effectiveness of crossover in simulated evolutionary optimization. , 1994, Bio Systems.

[5]  Larry J. Eshelman,et al.  On Crossover as an Evolutionarily Viable Strategy , 1991, ICGA.

[6]  Larry J. Eshelman,et al.  Crossover's Niche , 1993, ICGA.

[7]  Stuart A. Kauffman,et al.  The origins of order , 1993 .

[8]  H. Muhlenbein,et al.  Gene pool recombination and utilization of covariances for the Breeder Genetic Algorithm , 1995, Proceedings of 1995 IEEE International Conference on Evolutionary Computation.

[9]  Bernard Manderick,et al.  The Usefulness of Recombination , 1995, ECAL.

[10]  A. E. Eiben,et al.  Genetic algorithms with multi-parent recombination , 1994, PPSN.

[11]  A. E. Eiben,et al.  Orgy in the Computer: Multi-Parent Reproduction in Genetic Algorithms , 1995, ECAL.

[12]  William M. Spears,et al.  Crossover or Mutation? , 1992, FOGA.

[13]  Bernard Manderick,et al.  The Genetic Algorithm and the Structure of the Fitness Landscape , 1991, ICGA.

[14]  H. Mühlenbein,et al.  Gene Pool Recombination in Genetic Algorithms , 1996 .

[15]  P. Vittoz Man , 1962, Bloom.

[16]  Jim Smith,et al.  Recombination strategy adaptation via evolution of gene linkage , 1996, Proceedings of IEEE International Conference on Evolutionary Computation.

[17]  Thomas Bäck,et al.  An Overview of Evolutionary Algorithms for Parameter Optimization , 1993, Evolutionary Computation.