Application of self-organizing maps to genetic algorithms

This paper describes Self-Organizing Maps for Genetic Algorithm (SOM-GA). In this algorithm, the search performance of a real-coded genetic algorithm (RCGA) is enhanced with self-organizing map (SOM). The SOM is trained with the information of the individuals in the population. Sub-populations are generated from a whole population by the help of the map. The RCGA search is performed in the sub-populations. The Rastrigin function is considered as a test problem. The search performance of SOM-GA is compared with that of the RCGA. The results show that the use of the sub-population search algorithm improves the local search performance of the RCGA and therefore, SOM-GA can find better solutions in shorter CPU time than RCGA.