Project Description

In this research, we concentrate in spreading the population along the front by introducing the usage of the so-called pair-potential functions as density estimators in an external population (archive). By using the pair-potential functions as density estimators in an external population (archive), we show that:

• The MOEA is able to maintain solutions in the archive with high diversity for a longer period of time compared to the regular execution of the MOEA, and without meddling with its original definition.
• By studying the usage of several pair-potential functions in isolation, we provide insights into the performance dependence of a MOEA, and the pair-potential functions with respect to the shape of the Pareto front on different MOPs.