Integrated Radial Basis Function Networks with Adaptive Residual Subsampling Training Method for Approximation and Solving PDEs

Authors:
Hong Chen;Li Kong
Affiliations:
Department of Control Science and Engineering, Huazhong University of Science and Technology, Wuhan, China 430074 and Wuhan 2nd Ship Design and Research Institute, Wuhan, China 430064;Department of Control Science and Engineering, Huazhong University of Science and Technology, Wuhan, China 430074
Venue:
ISNN 2009 Proceedings of the 6th International Symposium on Neural Networks: Advances in Neural Networks - Part II
Year:
2009

Citing 6
Cited 0

Universal approximation using radial-basis-function networks

Neural Computation
Numerical solution of differential equations using multiquadric radial basis function networks

Neural Networks
Adaptive residual subsampling methods for radial basis function interpolation and collocation problems

Computers & Mathematics with Applications
Adaptive radial basis function methods for time dependent partial differential equations

Applied Numerical Mathematics
Integrated multiquadric radial basis function approximation methods

Computers & Mathematics with Applications
Multiobjective evolutionary optimization of the size, shape, and position parameters of radial basis function networks for function approximation

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a method for approximation and solving PDEs, based on integrated radial basis function networks (IRBFNs) with adaptive residual subsampling training algorithm. The Multiquadratic function is chosen as the transfer function of the neurons. The effectiveness of the method is demonstrated in numerical examples by approximating several functions and solving Burgers' equation. The result of numerical experiments shows that the IRBFNs with the proposed adaptive procedure requires less neurons to attain the accuracy than direct radial basis function (DRBF) network.