PROJECT TITLE :

A Multioperator Search Strategy Based on Cheap Surrogate Models for Evolutionary Optimization

ABSTRACT:

It is well-known that in evolutionary algorithms (EAs), totally different copy operators may be appropriate for different problems or in several running stages. To improve the algorithm performance, the ensemble of multiple operators has become popular. Most ensemble techniques achieve this goal by selecting an operator in line with a probability learned from the previous expertise. In contrast to these ensemble techniques, in this paper we tend to propose a low cost surrogate model-primarily based multioperator search strategy for evolutionary optimization. In our approach, a collection of candidate offspring solutions are generated by using the multiple offspring copy operators, and the simplest one in step with the surrogate model is chosen as the offspring solution. Two major benefits of this approach are: one) every operator can generate a resolution for competition compared to the likelihood-primarily based approaches and 2) the surrogate model building is comparatively low-cost compared to that in the surrogate-assisted EAs. The model is used to implement multioperator ensemble in 2 fashionable EAs, that's, differential evolution and particle swarm optimization. Thirty benchmark functions and the functions presented in the CEC 2013 are chosen because the check suite to evaluate our approach. Experimental results indicate that the new approach will improve the performance of single operator-based ways in the majority of the functions.