A surrogate-assisted hybrid optimization algorithm enhanced by opposition-based learning and its variants
Zhaoqi Ling · Journal of Physics Conference Series · 2021
Abstract Surrogate-Assisted Evolutionary Algorithms (SAEAs) are effective approaches to solve computationally expensive optimization problems by remarkably reducing real fitness evaluations. In this work, we proposed a surrogate-assisted hybrid optimization algorithm via combining a famous hierarchical SAEA Framework ESAO and a hybrid teaching-learning based optimization (TLBO) algorithm TLBO-SM. In addition, opposition-based learning (OBL) and its variants are used to enhance the global search ability. Experimental results on benchmark problems show that our method can outperform state-of-the-art SAEAs on most benchmark problems with the significant enhancement made by a recently proposed OBL variant.