Scheme processing, proportional selection, and the misallocation of trialsin genetic algorithms

Traditional selection in genetic algorithms has relied on reproduction in p roportion to observed fitness. There has been recent interest in assessing the result of proportional selection on schemata in the presence of random effects (e.g., noisy evaluation of solutions). The analysis presented here indicates, in contrast with previous literature, that the introduction of n oise to the evaluation of solutions can change the expected sampling of sch emata, even when the noise is zero mean, This effect is examined in a varie ty of settings. Unfortunately, this "misallocation of trials" can also resu lt simply from random initialization of a population. (C) 2000 Published by Elsevier Science Inc. All rights reserved.