C. Dachapak, S. Kanae, Z.J. Yang, and K. Wada (Japan)
Reproducing kernel Hilbert space, Orthogonal least squares algorithm, Radial basis function, Neural networks
The present study proposes a new radial basis function which is derived based on an idea of mapping data into a high dimensional feature space which is known as Repro ducing Kernel Hilbert Space (RKHS) and then performing Radial Basis Function (RBF) network in the feature space. Orthogonal Least Squares (OLS) method is employed to select a suitable set of centers (regressors) from a large set of candidates in order to obtain a sparse regression model in the feature space. The proposed method is employed to a scalar function approximation problem and a nonlinear system identiļ¬cation problem by simulations.
Important Links:
Go Back
