Migrants Selection and Replacement in Distributed Evolutionary Algorithms for Dynamic Optimization
Many distributed systems (task scheduling,moving priorities,mobile environments, ...) can be linked as Dynamic Optimization Problems (DOPs), since they require to pursue an optimal value that changes over time. We have focused on the utilization of Distributed Genetic Algorithms (dGAs), one of the domains still to be investigated for DOPs. A dGA essentially decentralizes the population in islands which cooperate through migrations of individuals. In this article, we analyze the effect of the migrants selection and replacement on the performance of dGAs for DOPs. Quality and distance based criteria are tested using a comprehensive set of benchmarks. Results show the benefits and drawbacks of each setting for DOPs.
Unable to display preview. Download preview PDF.
- 2.Branke, J., Kaussler, T., Schmidt, C., Schmeck, H.: A multi-population approach to dynamic optimization problems. In: 4th International Conference on Adaptive Computing in Design and Manufacture, pp. 299–308. Springer (2000)Google Scholar
- 3.Cantú-Paz, E.: Migration policies and takeover times in parallel genetic algorithms. In: Proc. of the GECCO. Morgan Kaufman (1999)Google Scholar
- 4.Homayounfar, H., Areibi, S., Wang, F.: An Island based GA for static/dynamic optimization problems. In: 3rd International DCDIS Conference on Engineering Applications and Computational Algorithms (2003)Google Scholar
- 7.Oppacher, F., Wineberg, M.: The shifting balance genetic algorithm: Improving the GA in a dynamic environment. In: Proc. of the Genetic and Evolutionary Computation Conference (GECCO), pp. 504–510. Morgan Kaufman (1999)Google Scholar
- 8.Park, T., Choe, R., Ryu, K.R.: Dual-population genetic algorithm for nonstationary optimization. In: Proc. of the GECCO, pp. 1025–1032. ACM (2008)Google Scholar
- 9.Ursem, R.K.: Multinational GAs: Multimodal optimization techniques in dynamic environments. In: Whitley, D., et al. (eds.) Proc. of the GECCO, pp. 19–26. Morgan Kaufmann (2000)Google Scholar