GENETIC ALGORITHMS - Soft Computing and Intelligent

The term genetic algorithm, almost universally abbreviated nowadays to GA, was first used by John Holland [1], whose book Adaptation in Natural and Aritificial Systems of 1975 was instrumental in creating what is now a flourishing field of research and application that goes much wider than the original GA. Many people now use the term evolutionary computing or evolutionary algorithms (EAs), in order to cover the developments of the last 10 years. However, in the context of metaheuristics, it is probably fair to say that GAs in their original form encapsulate

most of what one needs to know. Holland’s influence in the development of the topic has been very important, but several other scientists with different backgrounds were also involved in developing similar ideas. In 1960s Germany, Ingo Rechenberg [2] and Hans-Paul Schwefel [3] developed the idea of the Evolutionsstrategie (in English, evolution strategy), while— also in the 1960s—Bremermann, Fogel and others in the USA implemented their idea for what they called evolutionary programming. The common thread in these ideas was the use of mutation and selection—the concepts at the core of the neo-Darwinian theory of evolution. Although some promising results were obtained, evolutionary computing did not really take off until the 1980s. Not the least important reason for this was that the techniques needed a great deal of computational power. Neverthe- less, the work of these early pioneers is fascinating to read in the light of our current knowledge; David Fogel (son of one of the early pioneers) has documented some of this work in [4]. 1975 was a pivotal year in the development of genetic algorithms. It was in that year that Holland’s book was published, but perhaps more relevantly for those interested in metaheuristics, that year also saw the completion of a doctoral thesis by one of Holland’s graduate students, Ken DeJong [5]. Other students of Holland’s had completed theses 56 C. Reeves in this area before, but this was the first to provide a thorough treatment of the GA’s capabilities in optimization. A series of further studies followed, the first conference on the nascent subject was convened in 1985, and another graduate student of Holland’s, David Goldberg, produced first an award-winning doctoral thesis on his application to gas pipeline optimization, and then, in 1989, an influential book [6]—Genetic Algorithms in Search, Optimization, and Machine Learning. This was the final catalyst in set- ting off a sustained development of GA theory and applications that is still growing rapidly. Optimization has a fairly small place in Holland’s work on adaptive systems, yet the majority ofresearch on GAs tends to assume this is their purpose. DeJong, who initiated this interest in optimization, has cautioned that this emphasis may be misplaced in a paper [7] in which he contends that GAs are not really function optimizers, and that this is in some ways incidental to the main theme of adaptation. Nevertheless, using GAs for optimization is very popular, and frequently successful...

Website: 150.214.190.154 | Filesize: -
No of Page(s): 28
Download Chapter 3 GENETIC ALGORITHMS - Soft Computing and Intelligent ....pdf