In this paper, we introduce a new idea to reduce the hard time consuming task of finding the set of initial parameter values of an evolutionary algorithm that uses the inheritance strategy. The key idea is to adapt the parameter values during the execution of the algorithm. This adaptation is guided by the degree of difficulty shown by the problem, which is being solved, for the algorithm. This strategy has been tested using an evolutionary algorithm to solve CSPs, but can easily be extended to any evolutionary algorithm which uses inheritance. A set of benchmarks showed that the new strategy helps the algorithm to solve more problems than the original approach. © Springer-Verlag Berlin Heidelberg 2004.
CITATION STYLE
Bonnaire, X., & Riff, M. C. (2004). Adapting evolutionary parameters by dynamic filtering for operators inheritance strategy. In Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 3315, pp. 225–234). https://doi.org/10.1007/978-3-540-30498-2_23
Mendeley helps you to discover research relevant for your work.