Cooperative model for genetic operators to improve GAs

This work proposes a new empirical model that puts genetic operators in a cooperative stand with each other. Two parallel operators produce offspring and fulfil specific roles: self-reproduction with mutation (SRM) as a permanent source of diversity to induce the appearance of beneficial mutations and crossover and mutation (CM) to propagate them in the population. An extinctive selection mechanism subjects CM´s and SRM´s offspring to compete for survival and to guarantee the preservation of beneficial mutations for the next generation. SRM is implemented with an adaptive mutation schedule, which acts depending on SRM´s contribution to the actual population, to keep control of the exploration-exploitation balance. Two adaptive mutation schemes are investigated for SRM, adaptive dynamic segment (ADS) and adaptive dynamic probability (ADP). CM´s mutation is used to create the appropriate conditions in which SRM´s offspring could be competitive with CM´s offspring. Thus the expected cooperation between CM and SRM emerges resulting in higher convergence velocity and higher convergence reliability. The proposed model is tested with the 0/1 multiple knapsack NP-hard combinatorial optimization problem where it impressively outperforms a canonical genetic algorithm as well as two other enhanced GAs