Title :
Pareto RBF networks based on multiobjective evolutionary computation
Author :
Kondo, Nobuhiko ; Hatanaka, Toshiharu ; Uosaki, Katsuji
Author_Institution :
Dept. of Inf. & Phys. Sci., Osaka Univ., Japan
Abstract :
In this paper, multi-objective structure selection of RBF networks based on evolutionary computation is considered. The RBF network structures are encoded to the chromosomes in GA. Then, they evolve toward Pareto optimum defined by some objective functions concerning with model accuracy, model complexity and connection weights. Some numerical simulation results indicate that the proposed approach can give Pareto optimum structures.
Keywords :
Pareto optimisation; genetic algorithms; nonlinear systems; radial basis function networks; Pareto RBF network; chromosomes; multiobjective evolutionary computation; radial basis function networks;
Conference_Titel :
SICE 2004 Annual Conference
Conference_Location :
Sapporo
Print_ISBN :
4-907764-22-7
