Single and multiobjective evolutionary optimization assisted by Gaussian random field metamodels

摘要

Abstract: This paper presents and analyzes in detail anefficient search method based on Evolutionary Algorithms (EA)assisted by local Gaussian Random Field Metamodels (GRFM).It is created for the use in optimization problems with computationallyexpensive evaluation function(s). The role of GRFM isto predict objective function values for new candidate solutionsby exploiting information recorded during previous evaluations.Moreover, GRFM are able to provide estimates