Dynamically Adjusting Migration Rates for Multi-Population Genetic Algorithms

In this paper, the issue of adapting migration parameters for MGAs is investigated. We examine, in particular, the effect of adapting the migration rates on the performance and solution quality of MGAs. Thereby, we propose an adaptive scheme to adjust the appropriate migration rates for MGAs. If the individuals from a neighboring sub-population can greatly improve the solution quality of a current population, then the migration from the neighbor has a positive effect. In this case, the migration rate from the neighbor should be increased; otherwise, it should be decreased. According to the principle, an adaptive multi-population genetic algorithm which can adjust the migration rates is proposed. Experiments on the 0/1 knapsack problem are conducted to show the effectiveness of our approach. The results of our work have illustrated the effectiveness of self-adaptation for MGAs and paved the way for this unexplored area.