Asymmetric Pareto-adaptive Scheme for Multiobjective Optimization
A core challenge of Multiobjective Evolutionary Algorithms (MOEAs) is to attain evenly distributed Pareto optimal solutions along the Pareto front. In this paper, we propose a novel asymmetric Pareto-adaptive (apa) scheme for the identification of well distributed Pareto optimal solutions based on the geometrical characteristics of the Pareto front. The apa scheme applies to problem with symmetric and asymmetric Pareto fronts. Evaluation on multiobjective problems with Pareto fronts of different forms confirms that apa improves both convergence and diversity of the classical decomposition-based (MOEA/D) and Pareto dominance-based MOEAs (paε-MyDE).
KeywordsMultiobjective Optimization Hypervolume Pareto-adaptive
Unable to display preview. Download preview PDF.
- 4.Durillo, J.J., Nebro, A.J., Alba, E.: The jmetal framework for multi-objective optimization: Design and architecture. IEEE Congr. Evol. Comput., 1–8 (2010)Google Scholar