A modified genetic algorithm: meta-level control of migration in a distributed GA

An algorithm for meta-level control of migration in a distributed mode is presented to improve the performance of genetic algorithms. The evolutionary model of creature and an example are given first. If gives a good inspiration to the modified genetic algorithm. The key characteristic of the algorithm is that the meta-level control of migration on the low-level sub-genetic algorithm sets (meanwhile each of the sub-genetic algorithm sets has its own genetic algorithm running on). The comparison between the modified genetic algorithm and a standard genetic algorithm is given at the end of this paper. The results have shown the effectiveness of the modified genetic algorithm