This paper proposes an improvement of Evolutionary Strategies for objective functions with locally correlated variables. It focusses on detecting local dependencies among variables of the objective function on the basis of the current population and transforming the original objective function into a new one of a smaller number of variables. Such a transformation is updated in successive iterations of the evolutionary algorithm to reflect local dependencies over successive neighborhoods of optimal solutions. Experiments performed on some popular benchmark functions confirm that the improved algorithm outperforms the original one. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Lipinski, P. (2010). Evolution strategies for objective functions with locally correlated variables. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6283 LNCS, pp. 352–359). https://doi.org/10.1007/978-3-642-15381-5_43
Mendeley helps you to discover research relevant for your work.