Chromosome (evolutionary algorithm)

In evolutionary algorithms a chromosome, also called a genotype, is a set of parameters that together define one proposed solution to the problem the algorithm is trying to solve. The set of all such solutions is the population, and the genome of an individual consists of one or, more rarely, several chromosomes. A chromosome is built from genes, each holding one or more semantically connected parameters, often called decision variables, which determine or influence phenotypic characteristics of the individual. In the basic genetic algorithm the chromosome is a binary string, while later variants use many other data structures, including lists of integer or real values, permutations for ordering problems such as the traveling salesman tour, gene objects of dynamic length for complex scheduling tasks, and syntax trees in genetic programming. Designing a chromosome means choosing a representation and a genotype to phenotype mapping. A well suited design reaches every admissible point of the search space, avoids redundancy, keeps small changes in the chromosome causing small changes in the phenotype, and excludes prohibited regions, and it must be matched by suitable mutation and crossover operators.