NSGA-III with Reference Points Adaptation

• Rheidner Silva Universidade Federal do Sergipe
• André Britto Universidade Federal do Sergipe

DOI: https://doi.org/10.5753/eniac.2019.9312

Abstract

Multi-Objective Evolutionary Algorithms (MOEAs) face certain difficulties solving problems with more than 3 objective functions, called Many-Objective Optimization Problems (MaOPs). New algorithms emerged to circumvent this problem, including NSGA-III. This algorithm explores the concept of reference points to make the selection of the solutions. But, it still has certain limitations and can be improved. This paper proposes two algorithms to adapt the reference points of the NSGA-III, based on the MOEA/D-AWA. The algorithms were evaluated to verify if the proposed adaptation procedure improves the performance of the NSGA-III.

Keywords: Multiobjective Optmization, Reference Points Adaptation

References

Carvalho, M. and Britto, A. (2018). Influence of reference points on a many-objective optimization algorithm. In 2018 7th Brazilian Conference on Intelligent Systems (BRACIS), pages 31–36.

Das, I. and Dennis, J. (1998). Normal-boundary intersection: A new method for generating the pareto surface in nonlinear multicriteria optimization problems. SIAM Journal on Optimization, 8(3):631–657.

Deb, K. and Jain, H. (2014). An evolutionary many-objective optimization algorithm using reference-point-based nondominated sorting approach, part i: Solving problems with box constraints. IEEE Transactions on Evolutionary Computation, 18(4):577– 601.

Deb, K., Pratap, A., Agarwal, S., and Meyarivan, T. (2002). A fast and elitist multiobjective genetic algorithm: Nsga-ii. IEEE Transactions on Evolutionary Computation, 6(2):182–197.

Qi, Y., Ma, X., Liu, F., Jiao, L., Sun, J., and Wu, J. (2014). Moea/d with adaptive weight adjustment. Evol. Comput., 22(2):231–264.

Schutze, O., Lara, A., and Coello, C. A. C. (2011). On the influence of the number of objectives on the hardness of a multiobjective optimization problem. IEEE Transactions on Evolutionary Computation, 15(4):444–455.

Zhang, Q. and Li, H. (2007). Moea/d: A multiobjective evolutionary algorithm based on decomposition. IEEE Transactions on Evolutionary Computation, 11(6):712–731.