A Pareto-based Symbiotic Relationships Model for Unconstrained Continuous Optimization

Abstract

Symbiotic relationships are one of several phenomena that can be observed in nature. These relationships consist of interactions between organisms and can lead to benefits or damages to those involved. In an optimization context, symbiotic relationships can be used to perform information exchange between populations of candidate solutions to a given problem. This paper presents an information exchange model inspired by symbiotic relationships and applies the model to unconstrained single-objective continuous optimization problems. The symbiotic relationships are modelled using the Pareto dominance criteria inside a computational ecosystem for optimization. The Artificial Bee Colony algorithm is used to compound the populations of the ecosystem. Four models of relationships are analyzed: slavery, competition, altruism and mutualism. Thirty unconstrained single-objective continuous benchmark functions with high number of dimensions (d = 200) are tested and obtained results compared. Results suggest that the proposed model for information exchange favors the balance between exploration and exploitation leading to better results.