Improving the NSGA-II Performance with an External Population
The NSGA-II algorithm is among the best performing ones in the area of multiobjective optimization. The classic version of this algorithm does not utilize any external population. In this work several techniques of reintroducing specimens from the external population back to the main one are proposed. These techniques were tested on multiobjective optimization problems named ZDT-1, ZDT-2, ZDT-3, ZDT-4 and ZDT-6. Algorithm performance was evaluated with the hypervolume measure commonly used in the literature. Experiments show that reintroducing specimens from the external population improves the performance of the algorithm.
KeywordsMultiobjective optimization NSGA-II External population
- 4.Purshouse, R.: On the evolutionary optimisation of many objectives. Ph. D. thesis, The University of Sheffield, UK (2003)Google Scholar
- 5.Sharma, D., Kumar, A., Deb, K., Sindhya, K.: Hybridization of SBX based NSGA-II and sequential quadratic programming for solving multi-objective optimization problems. In: IEEE Congress on Evolutionary Computation, pp. 3003–3010. IEEE (2007)Google Scholar
- 8.Yang, L., Yang, C., Liu, Y.: Particle swarm optimization with simulated binary crossover. In: 2014 Fifth International Conference on Intelligent Systems Design and Engineering Applications (ISDEA), pp. 710–713 (2014)Google Scholar
- 10.Zitzler, E., Laumanns, M., Thiele, L.: SPEA2: Improving the strength pareto evolutionary algorithm for multiobjective optimization. In: Giannakoglou, K., et al. (eds.) Evolutionary Methods for Design, Optimisation and Control with Application to Industrial Problems (EUROGEN 2001), pp. 95–100. International Center for Numerical Methods in Engineering (CIMNE) (2002)Google Scholar