Boosting the performance of hybrid Nature-Inspired algorithms: Application from the financial optimization domain

In most optimization problems, the solution space is very vast. At the same time, the imposed constraints and limitations provide additional burdens for the underline algorithm. Although hybrid intelligent approaches have proven their potential in demanding problem settings, in most cases, they manage to approximate a good-quality solution, meaning that there is room for improvement. Based on preliminary findings from previous studies, the aim of this paper is to highlight the impact of the incorporation of heuristic information, i.e. expert’s knowledge etc., to the overall performance of a hybrid Nature-Inspired algorithm. Experimental findings support the related results from HAIS 2017 conference [Tzanetos, Vassiliadis and Dounias, 2017, Lecture Notes in Computer Science, 10334]. More specifically, under the framework of a financial portfolio optimization problem, the heuristic information-enhanced hybrid algorithm manages to reach a new optimal solution. What is more, in the scope of Nature-Inspired algorithms, some preliminary results, under the framework of hybrid algorithms, from a newly proposed metaheuristic are provided, though further experimentation and fine-tuning are required.