Local Meta-models for Optimization Using Evolution Strategies


S. Kern, N. Hansen, P. Koumoutsakos, Lecture Notes in Computer Science, 4193, 939-948, Springer Berlin, 2006



We employ local meta-models to enhance the efficiency of evolution strategies in the optimization of computationally expensive problems. The method involves the combination of second order local regression meta-models with the Covariance Matrix Adaptation Evolution Strategy. Experiments on benchmark problems demonstrate that the proposed meta-models have the potential to reliably account for the ranking of the offspring population resulting in significant computational savings. The results show that the use of local meta-models significantly increases the efficiency of already competitive evolution strategies.