Evolutionary Computing

Evolutionary Computing

Thomas E. Potok (Oak Ridge National Laboratory, USA), Xiaohui Cui (Oak Ridge National Laboratory, USA) and Yu Jiao (Oak Ridge National Laboratory, USA)
DOI: 10.4018/978-1-59904-982-3.ch008
OnDemand PDF Download:
$37.50

Abstract

The rate at which information overwhelms humans is significantly more than the rate at which humans have learned to process, analyze, and leverage this information. To overcome this challenge, new methods of computing must be formulated, and scientist and engineers have looked to nature for inspiration in developing these new methods. Consequently, evolutionary computing has emerged as new paradigm for computing, and has rapidly demonstrated its ability to solve real-world problems where traditional techniques have failed. This field of work has now become quite broad and encompasses areas ranging from artificial life to neural networks. This chapter specifically focuses on two sub-areas of nature-inspired computing: Evolutionary Algorithms and Swarm Intelligence.
Chapter Preview
Top

Evolutionary Algorithms

Charles Darwin radically changed the way evolutionary biology is viewed in his work entitled “Origin of Species” published in 1859 (Darwin, 1859). In this work, Darwin describes his theory of natural selection based on his experience and observations of nature around the world. Darwin states that there is an implicit struggle for survival because of species producing more offspring than can grow to adulthood and that food sources are limited. Because of this implicit struggle, sexually reproducing species create offspring that are genetic variants of the parents. Darwin theorizes that it is this genetic variation that enables some offspring to survive in a particular environment much better than other offspring with different genetic variations. As a direct result of this “enhanced” genetic variation, these offspring not only survive in the environment, but go on to reproduce new offspring that carry some form of this enhanced genetic variation. In addition, those offspring that are not as suited for the environment do not pass on their genetic variation to offspring, but rather die off. Darwin then theorizes that over many generations of reproduction, new species that are highly adapted to their specific environments will emerge. It is this theory of natural selection that forms the theoretical foundation for the field of Evolutionary Algorithms (EA).

Following in the footsteps of Darwin, John Holland dramatically altered the computer science and artificial intelligence fields in 1975 with his publication entitled “Adaptation in Natural and Artificial Systems” (Holland 1975). In this work, Holland describes a mathematical model for the evolutionary process of natural selection, and demonstrates its use in a variety of problem domains. This seminal work by Holland created the fertile soil by which the field of Evolutionary Algorithms grew and thrived. In the same year and under the direction of Holland, Ken De Jong’s dissertation entitled “An Analysis of the Behavior of a Class of Genetic Adaptive Systems” helps fully demonstrate the possibilities of using evolutionary algorithms for problem solving (De Jong, 1975). In 1989, the field of evolutionary algorithms received a fresh injection of enthusiasm with the publication of David Goldberg’s work entitled “Genetic Algorithms in Search, Optimization, and Machine Learning” (Goldberg, 1989). The momentum of development continued with Melanie Mitchell’s 1996 work entitled “An Introduction to Genetic Algorithms,” which helped to further solidify the theoretical foundations of EAs (Mitchell, 1996). Ever since then, the field has continued to grown and the practical applications of EA’s are abounding with success stories (Chambers, 2000; Coley, 2001; Haupt, 1998).

Complete Chapter List

Search this Book:
Reset