In this paper we derive theoretical expressions to compute expected population variance for Differential Evolution (DE) variants - DE/best/1/bin, DE/rand/2/bin and DE/best/2/bin by directly extending Zaharie's work on DE/rand/1/bin. The study includes comparing the theoretical and empirical evolution of population variance of three DE variants. This work provides insight about the explorative power of the variants and explains their behavior. © 2010 Springer-Verlag.
CITATION STYLE
Jeyakumar, G., & Velayutham, C. S. (2010). A comparative study on theoretical and empirical evolution of population variance of differential evolution variants. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6457 LNCS, pp. 75–79). https://doi.org/10.1007/978-3-642-17298-4_7
Mendeley helps you to discover research relevant for your work.