Genetic algorithms apply the biological principles of selection, mutation, and crossover to a population set containing individuals representing target solutions to a given problem. Using these principles genetic algorithms attempt to create a migration of the individuals in subsequent generations toward the optimal solution.
This project is an attempt to visually represent the progress of a genetic algorithm. The coordinate fitness program attempts to find the maximum or minimum value of a given function. It visually represents the progress of the algorithm by providing a plot of each individual in each generation in time. It is then possible to view the migration of points toward a known maximum or minimum value. Visual representation is also achieved by a plot of the highest and lowest fitness per generation, as well as average fitness per generation.
The parameters of crossover rate and mutation rate can be altered. This allows experimentation in finding a good combination of these rates for a particular function, and viewing the results. Many involved in the field of genetic algorithms believe that this is an area of the subject that requires further research.
Available by permission of the author. Reproduction or retransmission of this material in any form is prohibited without expressed written permission of the author.
Jones, Paul W., "Genetic Algorithms: A Visual Search" (1997). Honors Theses. 584.