Evolutionary computation for discovery
Kenneth Alan De Jong · Communications of the ACM · 1999
It sounds a bit strange to use the words "evolution"and "discovery" together.We think of evolution as a slow, incremental process of gradual change and discovery more in terms of abrupt "Eureka!" events.And yet there is at least one arena in which these two notions fit hand-in-glove: the Evolutionary Computation for DiscoveryKenneth A. De Jong New automated tools hold the promise of allowing users to find more, learn more, and build more opportunities.ANGELA WILCZYNSKI/WIZZLE.SIMPLENET.COM/FRACTALS/FRACTALINTRO.HTM use of evolutionary computation techniques to facilitate and automate important aspects of discovery in science and engineering.Space constraints do not permit a comprehensive survey of the wide variety of fascinating work in this area.Rather, this article will simply illustrate the use of evolutionary techniques for discovery by looking at the details of a few interesting examples.The term "evolutionary computation" refers to the use of evolutionary algorithms to solve difficult computational problems.Genetic Algorithms (GAs) are the most well known and most widely