Title :
The Comparison of Two Kinds of Reproduction in Multiobjective Optimization
Author :
Yao, Xue-fei ; Liu, Hai-Lin
Author_Institution :
Fac. of Appl. Math., Guangdong Univ. of Technol., Guangzhou, China
Abstract :
Most of constraint handling papers have focused on the selection of individuals by trade-off the feasible and infeasible regions. This paper studies the effect of two kinds of reproduction in constraint multiobjective optimization. It compares a probabilistic model-based multiobjective evolutionary algorithm to a genetic algorithm. They all use a min-max selection strategy as the main frame structure. The comparison with 19 constraint multiobjective optimization testing instances is given in this paper. The comparison of the instances can show us which reproduction should be used to get a better result when facing different cases.
Keywords :
constraint theory; genetic algorithms; minimax techniques; probability; constraint handling; constraint multiobjective optimization; genetic algorithm; min-max selection; probabilistic model-based multiobjective evolutionary algorithm; reproduction strategy; constraint; estimation of distribution algorithm; multiobjective optimization;
Conference_Titel :
Computational Intelligence and Security (CIS), 2010 International Conference on
Conference_Location :
Nanning
Print_ISBN :
978-1-4244-9114-8
Electronic_ISBN :
978-0-7695-4297-3
DOI :
10.1109/CIS.2010.28
