Genetic Algorithms + Data Structures = Evolution Programs PDF

'What does your Master teach?' asked a visitor. 'Nothing,' said the disciple. 'Then why does he give discourses?' 'He only points the way - he teaches nothing.' Anthony de Mello, One Minute Wisdom During the last three decades there has been a growing interest in algorithms which rely on analogies to natural processes.

The emergence of massively par- allel computers made these algorithms of practical interest.

The best known algorithms in this class include evolutionary programming, genetic algorithms, evolution strategies, simulated annealing, classifier systems, and neural net- works.

Recently (1-3 October 1990) the University of Dortmund, Germany, hosted the First Workshop on Parallel Problem Solving from Nature [164].

This book discusses a subclass of these algorithms - those which are based on the principle of evolution (survival of the fittest).

In such algorithms a popu- lation of individuals (potential solutions) undergoes a sequence of unary (muta- tion type) and higher order (crossover type) transformations.

These individuals strive for survival: a selection scheme, biased towards fitter individuals, selects the next generation.

After some number of generations, the program converges - the best individual hopefully represents the optimum solution.

There are many different algorithms in this category.

To underline the sim- ilarities between them we use the common term "evolution programs" .