Evolutionary Algorithms - Self-Adaptive Heuristics for Evolutionary Computation - page 7

2 Evolutionary Algorithms
This chapter presents the foundations of the EC framework. After the classifica-
tion into the methods of computational intelligence and the definition of terms
like evolutionary convergence, the basic concepts of EAs are presented. This
introduction contains representations, genetic operators and concepts like popu-
lation models, fitness landscapes and termination conditions. Then the chapter
gives an overview of special types of EAs. The following section 2.2 introduces
the ( μ/ρ + ,λ ) - ES, which is the basis of most evolutionary concepts of this work.
A survey of theoretical approaches to EC completes this chapter.
2.1 Introduction to Evolutionary Computation
EAs are biologically inspired, population based, randomized search meta-
heuristics. They exhibit the fundamental principles of biological evolution: in-
heritance of genes, variation of genes in a population, translation of genotype
into phenotype and selection of the fittest phenotype in the sense of the Dar-
winian principle of survival of the fittest. In the sixties and seventies Fogel [44],
Holland [58], Rechenberg [114] and Schwefel [133] translated this paradigm of
evolution into a concept of algorithms that is called EC today. It has grown to
a rich and frequently used optimization method. It comprises several variants of
algorithms which are structurally very similar, but are specialized to the search
domain characteristics.
ES are one of the four main variants of EC techniques. They were invented by
Rechenberg [113] and Schwefel [131] in the sixties in Germany and are usually
associated with continuous evolutionary optimization. They exhibit operators
which are appropriate to real valued search spaces. A more detailed introduction
is given later in this chapter. First of all, the chapter starts with an introduction
into CI and the EC framework.
2.1.1 Computational Intelligence
The main goal of artificial intelligence (AI) is the construction of systems which
are able to learn and behave intelligently. Pioneers like John von Neumann