Understanding Genetic Algorithms

    Genetic algorithms closely resemble the biological model of chromosomes and genes. Individual organisms in a genetic algorithm generally consist of a single chromosome. These chromosomes are composed of genes. By manipulating the genes, new chromosomes are created, which possess different traits. These manipulations occur through crossover and mutation, just as they occur in nature. Crossover is analogous to the biological process of mating, and mutation is one way in which new information can be introduced into an existing population.

Understanding Genes

    In a genetic algorithm, genes represent individual components of a solution. This is a very important concept in the analysis of a problem for which a genetic algorithm is to be used. To effectively solve a problem, you must determine a way to break it into its related components, or genes. The genes will then be linked together to form a chromosome. Life forms in this simulation consist of a single chromosome; therefore, each chromosome will represent one possible solution to a problem.

    Later in this chapter, we will examine the traveling salesman problem. You will be shown how to decompose the solution for this problem into individual genes. Additionally, in this chapter you will be shown how to make the individual weights in a neural network represent the genes in the chromosome.

    It is important to note that there is a finite number of genes that are used. Individual genes are not modified as the organisms evolve. It is the chromosomes that evolve when the order and makeup of their genes are changed.

Understanding Chromosomes

    As explained above, each organism in our genetic algorithm contains one chromosome. As a result, each chromosome can be thought of as an “individual” or a solution composed of a collection of parameters to be optimized. These parameters are genes, which you learned about in the previous section. Each chromosome is initially assigned a random solution or collection of genes. This solution is used to calculate a “fitness” level, which determines the chromosome’s suitability or “fitness” to survive—as in Darwin’s theory of natural selection. If a chromosome has a high level of “fitness,” it has a higher probability of mating and staying alive.