Genetic global optimization algorithms
Sergej M. Ermakov, Dmitriy N. Semenchikov · Communications in Statistics - Simulation and Computation · 2019
Two new facts regarding genetic extremum search algorithms are presented. The first one, which is based on the use of a modification of the simulated annealing method, makes it possible to distinguish close extrema at the initial stage and indicates the strategy of behavior in the case of many equal extrema. The second fact (a lemma and its consequence) points out new methods for implementing the evolution of the covariance matrix. Numerical examples are given.