Cellular Automata and Fractal Evolution

Genetic Algorithms and Artificial Evolution - Crossover

Crossover is a method of recombining two strings of genetic material.
Simplest is the single-point crossover: snip both strings at the same location and interchange the pieces.
Another method, multiple-point crossover, involves recombining the strings at several locations. We shall consider only single-point crossovers.
First here is an illustration of single-point corssover using two sentences.
Click the picture to animate.
Here is an example using the CA classifier system:
Here are patterns generated by the parents, both from the same random initial distribution. Click on the small picture for a larger version in a new window.
Here are patterns generated by the children, both from the same random initial distribution. Click on the small picture for a larger version in a new window.
Crossover allows large excursions in genotype space. It is aided by Holland's Schema theorem: combinations of genes that increase fitness tend to be preserved and amplified by crossover in large populations.

Return to Genetic Algorithms and Artificial Evolution.