A Global Optimization Method of Radial Basis Function Networks for Function Approximation


The KIPS Transactions:PartB , Vol. 14, No. 5, pp. 377-382, Oct. 2007
10.3745/KIPSTB.2007.14.5.377,   PDF Download:

Abstract

This paper proposes a training algorithm for global optimization of the parameters of radial basis function networks. Since conventional training algorithms usually perform only local optimization, the performance of the network is limited and the final network significantly depends on the initial network parameters. The proposed hybrid simulated annealing algorithm performs global optimization of the network parameters by combining global search capability of simulated annealing and local optimization capability of gradient-based algorithms. Via experiments for function approximation problems, we demonstrate that the proposed algorithm can find networks showing better training and test performance and reduce effects of the initial network parameters on the final results.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
J. S. Lee and C. H. Park, "A Global Optimization Method of Radial Basis Function Networks for Function Approximation," The KIPS Transactions:PartB , vol. 14, no. 5, pp. 377-382, 2007. DOI: 10.3745/KIPSTB.2007.14.5.377.

[ACM Style]
Jong Seok Lee and Cheol Hoon Park. 2007. A Global Optimization Method of Radial Basis Function Networks for Function Approximation. The KIPS Transactions:PartB , 14, 5, (2007), 377-382. DOI: 10.3745/KIPSTB.2007.14.5.377.