ARTST 102

Michael Godwin

Genetic Algorithm

Genetic Algorithms are a subset of Evolutionary Algorithms that utilize techniques inspired by evolutionary biology to "evolve" solutions to a given problem.

The algorithm begins with a random population of chromosomes -- abstract representations of solutions. Next a fitness test is applied to each chromosome and a new population of solutions is created based on each chromosome's fitness; fitness is determined by how well that solution solves the initial problem. The precise method of choosing each new population is the secret to a successful implementation of Genetic Algorithm. A number of methods that mimic evolutionary biology are used to diversify each successive generation including cross-over, mutation, and elitism.

Evolutionary computing was invented in the 1960's by Ingo Rechenberg. John Holland further developed these ideas to invent the Genetic Algorithm in the 1970's. The University of Illinois hosted the first International Conference on Genetic Algorithms in the mid-80's and in 1989 the first commercially available genetic algorithm software became available. Today Genetic Algorithms are used frequently to determine solutions for a variety of problems.

Introduction to Genetic Algorithms at Czech Technical University
Bibliography: Wikipedia's Genetic Algorithm entry
Politics & Applications: Genetic Algorithms and Intelligent Design
International Society for Genetic and Evolutionary Computation
Illinois Genetic Algorithms Laboratory