论文信息 - Comparison of radial basis function approximation techniques

Comparison of radial basis function approximation techniques

This paper compares three different radial basis function neural networks, as well as the diffuse element method, according to their ability of approximation. This is very useful for the optimization of electromagnetic devices. Tests are done on several analytical functions and on the TEAM workshop problem 25.

Jean-Louis Coulomb | Avenir Kobetski | M. C. Costa | Yves Mare´chal | Ulf Jo¨nsson

[1] F. Girosi,et al. Networks for approximation and learning , 1990, Proc. IEEE.

[2] N. Takahashi,et al. Investigation of simulated annealing method and its application to optimal design of die mold for orientation of magnetic powder , 1996 .

[3] M. Repetto,et al. A multiquadrics-based algorithm for the acceleration of simulated annealing optimization procedures , 1996 .

[4] James J. Carroll,et al. Approximation of nonlinear systems with radial basis function neural networks , 2001, IEEE Trans. Neural Networks.

[5] Andrew R. Webb,et al. Shape-adaptive radial basis functions , 1998, IEEE Trans. Neural Networks.

[6] Jean-Louis Coulomb,et al. Diffuse-element method and quadtrees: two "ingredients" for an adaptive response surface , 2002 .