Evolutionary Algorithms: From Recombination to Search Distributions

Abstract

First we show that all genetic algorithms can be approximated by an algorithm which keeps the population in linkage equilibrium. Here the genetic population is given as a product of univariate marginal distributions. We describe a simple algorithm which keeps the population in linkage equilibrium. It is called the univariate marginal distribution algorithm (UMDA). Our main result is that UMDA transforms the discrete optimization problem into a continuous one defined by the average fitness W(p1, . . . , p n ) as a function of the univariate marginal distributions p i. For proportionate selection UMDA performs gradient ascent in the landscape defined by W(p). We derive a difference equation for p i which has already been proposed by Wright in population genetics. We show that UMDA solves difficult multimodal optimization problems. For functions with highly correlated variables UMDA has to be extended. The factorized distribution algorithm (FDA) uses a factorization into marginal and conditional distributions. For decomposable functions the optimal factorization can be explicitly computed. In general it has to be computed from the data. This is done by LFDA. It uses a Bayesian network to represent the distribution. Computing the network structure from the data is called learning in Bayesian network theory. The problem of finding a minimal structure which explains the data is discussed in detail. It is shown that the Bayesian information criterion is a good score for this problem.