Genetic algorithm (GA)

also canonical GA, variant of EA, which, (in general) in analogy to the biological DNA alphabet, operates on strings, usually bit strings of constant length (see, however, real-coded GA). The string corresponds to the genotype of the individual. The phenotype of the individual is realized by a mapping onto the object parameters (genotype-phenotype mapping). The fitness of the individual is usually a function of the objective function (to be optimized), which depends on the object parameters. The GA is usually characterized by fitness-proportionate selection and tournament selection, respectively. Crossover is the main variation operator; mutation (bit mutation) is usually regarded as a background operator or is not employed.