evolutionary algorithm

(redirected from Artificial evolution)
Also found in: Dictionary.

evolutionary algorithm

(EA) An algorithm which incorporates aspects of natural selection or survival of the fittest. An evolutionary algorithm maintains a population of structures (usually randomly generated initially), that evolves according to rules of selection, recombination, mutation and survival, referred to as genetic operators. A shared "environment" determines the fitness or performance of each individual in the population. The fittest individuals are more likely to be selected for reproduction (retention or duplication), while recombination and mutation modify those individuals, yielding potentially superior ones.

EAs are one kind of evolutionary computation and differ from genetic algorithms. A GA generates each individual from some encoded form known as a "chromosome" and it is these which are combined or mutated to breed new individuals.

EAs are useful for optimisation when other techniques such as gradient descent or direct, analytical discovery are not possible. Combinatoric and real-valued function optimisation in which the optimisation surface or fitness landscape is "rugged", possessing many locally optimal solutions, are well suited for evolutionary algorithms.
References in periodicals archive ?
Moreover, artificial evolution is establishing itself as a powerful technique for image synthesis and potentially for model synthesis.
Indeed, this particular exhibition is a clever complement to 1992's "Post Human," in which curator Jeffrey Deitch put forth the tech-savvy notion that the rush of modern life and its body-changing arts (plastic surgery and gene splicing among them) are fast-forwarding us into "a bold realm of artificial evolution.
Another approach is artificial evolution in microorganisms, in which thousands of generations of selective response in huge populations can be tracked, and a correspondingly large range of mutational possibilities tested.
About a year ago Stephanie Forrest of the University of New Mexico and Terry Jones of the Santa Fe Institute - the nerve center of "complexity" research - were fooling around with ECHO, an artificial evolution program developed by Chris Langton, the father of "artificial life.
Colfax's Advanced Design of GPU-Focused, 10,136-Core Parallel Computing Facility to Further the Artificial Evolution of Neural Circuitry
Because it is not possible to conduct artificial evolution with social insects we will use a new experimental system consisting of colonies of small mobile robots with simple vision and communication abilities.
The techniques rely on guided trial-and-error strategies inspired by Charles Darwin's theory of evolution by natural selection and known as evolutionary algorithms or artificial evolution (SN: 7/23/94, p.
As a generic example of artificial evolution we consider genetic algorithms.
Chapter 2 deals with the nature of selection on single characters in more complex systems, with very detailed reviews of artificial selection experiments for a variety of traits and of the artificial evolution of novel metabolic functions in bacteria, with a much briefer account of some of the classic studies of selection in the wild.
Artificial evolution can produce bizarre circuits that work--but do we need to understand them?