Empirical Study: Initial Population Diversity and Genetic Algorithm Performance

This article presents an empirical study regarding the hypothesis that higher diversity in initial populations for Genetic Algorithms can reduce the number of iterations required to reach an optimum and potentially increase solution quality. We develop the empirical study using some theoretical functions addressed by other researchers such that the input to the Genetic Algorithm is populations of differing diversity. It is expected that the effort in analyzing the initial population with a diversity measure is going to be compensated for by reducing the number of iterations required and perhaps improving solution quality.

• Type: Conference
• Year: 2007
• Where: AIPRF
• Authors: Pedro A. Diaz-Gomez, Dean F. Hougen
• Keywords: AIPRF 2007, Artificial Intelligence, Empirical Study, Genetic Algorithm, Initial Population