On the Benefits of Random Memorizing in Local Evolutionary Search

Abstract

For the calibration of laser induced plasma spectrometers robust and efficient local search methods are required. Therefore, several local optimizers from nonlinear optimization, random search and evolutionary computation are compared. It is shown that evolutionary algorithms are superior with respect to reliability and efficiency. To enhance the local search of an evolutionary algorithm a new method of random memorizing is introduced. It leads to a substantial gain in efficiency for a reliable local search.