Abstract
Genetic Algorithms (GA) are a common probabilistic optimization method based on the model of natural evolution. One important operator in these algorithms is the selection. Some works has been done to classify the different selection schemes as roulette wheel selection, tournament selection etc. An enhanced version of tournament selection named elite tournament selection is introduced in this paper. This novel selection method solves probably the only one disadvantage of the standard tournament selection, which is that it does not guarantee reproduction of the best solution. In the part of this paper probability equations for the tournament and elite tournament selection are defined. On this base we derive further conclusions. The binomial distribution and convolution are used for mathematical description. Theoretical calculations are verified by means of real experiments.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call