Theory And Practice On Some Evolutionary Algorithms

In this paper, the theory and practice of Evolutionary Computationhave been systematically introduced. We focus our research on somenewly and important Evolutionary Algorithms, as well as theirapplication in function optimization, symbolic regression and MKPproblem.Estimation of Distribution Algorithms (EDAs) were first introducedby Muhlenbein and Paab in 1996. In contrast to traditional evolutionarytechnology, EdAs reproduce the next population through the jointprobability distribution associated with the individuals of variablesselected at each generation. The probability distribution is calculatedfrom a database of selected individuals of previous generation. Arelaxed complemental mechanism was coordinated into EDAs in ourresearch.Gene Expression Programming (GEP) was first proposed byCandida Ferreir in 2001. Expressions different in size and shape wasencoding in a linear struct in GEP. Its application include digital circuitdesign, data mining etc. We proposed a parallel framework of GEP inthis paper.