Increasing speed of genetic algorithm-based instance selection

We discuss several ways to accelerate genetic algorithm-based instance selection, where the two objectives are a minimal number of training instances and maximal accuracy of the classifier (we use neural networks) on the test data. We discuss several ways to accelerate the process, but we especially focus on two parameters: fitness function and chromosome length reduction. We evaluate different fitness functions, discuss their performance and propose the guidance for choosing the optimal one in respect to the process speed and stability. We also discuss the possibility of reducing the chromosome length during the optimization by excluding from further optimization these positions that are unlikely to change. We verify our method experimentally as well on several real-world datasets.