A New Multi-Objective Driving-Training-Based Optimization Algorithm

This paper presents a driving-training-based optimization algorithm for multi-objective problems. In this first multi-objective variant, each learner driver participates in two phases. A set of driving instructors is adaptively selected based on the size of the population and the rate of progress in the optimization process. In the first phase, the learner driver is trained by a driving instructor. The learner driver then has the option of modeling the driving instructor's skills or improving its performances through personal practice in the second phase. The proposed algorithm is compared to well-known and state-of-the-art algorithms. It has demonstrated promising results in multi-objective problems.