Title :
Study of RBF Neural Network Based on Improved OLS Algorithm
Author :
Zhezhao, Zeng ; Jie, Jiang
Author_Institution :
Coll. of Electr. & Inf. Eng., Changsha Univ. of Sci. &Technol., Changsha, China
Abstract :
Training of parameters in RBF neural network, this article proposes an optimization of RBF neural network parameters algorithm, which can overcome the disadvantages of select of data center and weights in RBF neural network. The algorithm process input data normalization and compute network output and hidden layer output angle cosine firstly, a set of data being established as the network center when cosine value is most minimum. And then determine network weights based on OLS algorithm. Simulation results show that the algorithm can reduce the training sample data and increase network training speed when train RBF neural network.
Keywords :
learning (artificial intelligence); least squares approximations; optimisation; radial basis function networks; RBF neural network parameter training; data center; data normalization; hidden layer output angle cosine computation; improved OLS algorithm; network center; orthogonal least square algorithm; radial basis function neural network; Artificial neural networks; Classification algorithms; Least squares approximation; Neurons; Radial basis function networks; Simulation; Training; Adaptive Learning; Cosine Method; OLS Algorithm; RBF Neural Network;
Conference_Titel :
Intelligent Computation Technology and Automation (ICICTA), 2011 International Conference on
Conference_Location :
Shenzhen, Guangdong
Print_ISBN :
978-1-61284-289-9
DOI :
10.1109/ICICTA.2011.70