Abstract The problem of training RBF (Radial Basis Function) neural network for pattern recognition is considered. In this paper, taking account of the specific feature of classification problem, a new training algorithm based on the regional mapping and novelty condition of RAN (Resource Allocating Network) is proposed. The results show the effectiveness of the proposed approach in RBF network training for pattern recognition, mainly in shortening the learning time, simplifying the structure of network and improving the classification accuracy.
[1] Jin C. Theory Analysis for Error Function to Feedforward Neural Networks with Noise. Journal of Computer Research and Develop-
ment, 2002, 39(2): 213-216 (in Chinese) (金 聪.含噪声前馈神经网络误差函数的理论分析.计算机研究与发展, 2002, 39(2): 213-216) [2] Lampariello F, Sciandrone M. Efficient Training of RBF Neural Networks for Pattern Recognition. IEEE Trans on Neural Networks, 2001, 12(5): 1235-1242 [3] Jiang L F, Liu B, Shi L H. A Improved BP Algorithm and Applied in the Pattern Recognition. Journal of Harbin University of Science and Technology, 2003, 8(3):90-93 (in Chinese) (姜立芳,刘 泊,施莲辉.一种改进的BP算法及其在模式识别中的应用.哈尔滨理工大学学报, 2003, 8(3): 90-93) [4] Jin C. Structure Modality of the Error Function for Feedforward Neural Networks. Journal of Computer Research and Development, 2003, 39(7): 913-917 (in Chinese) (金 聪.前馈神经网络误差函数的结构形式.计算机研究与发展, 2003, 39(7): 913-917) [5] Yang Y H, Li X, Jiang F Z. Freeway Incident Detection Based on OLS and RBF Neural Networks. Journal of System Simulation, 2003, 15(5): 709-712 (in Chinese) (杨耀华,李 昕,江芳泽.基于OLS算法的RBF神经网络高速公路事件探测.系统仿真学报, 2003, 15(5): 709-712) [6] Sun J, Shen R M, Han P. An Original RBF Network Learning Algorithm. Chinese Journal of Computers, 2003, 26(11): 1562-1567 (in Chinese) (孙 健,申瑞民,韩 鹏.一种新颖的径向基函数(RBF)网络学习算法.计算机学报, 2003, 26(11): 1562-1567) [7] Sarimveis H, Alexandridis A, Mazarakis S. A New Algorithm for Developing Dynamic Radial Basis Function Neural Network Models Based on Genetic Algorithms. Computers and Chemical Engineering, 2004, 28(1-2): 209-217 [8] Liao Y, Fang S C, Henry L W. Relaxed Conditions for Radial-Basis Function Networks to be Universal Approximators. Neural Networks, 2003, 16(7): 1019-1028 [9] Ghodsi A, Schuurmans D. Automatic Basis Selection Techniques for RBF Networks. Neural Networks, 2003,16(5-6): 809-816 [10] Wei H K, Ding W M, Song W Z , Xu S X. Dynamic Method for Designing RBF Neural Networks. Control Theory and Applications, 2002, 19(5): 673-680 (in Chinese) (魏海坤,丁维明,宋文忠,徐嗣鑫.RBF网的动态设计方法.控制理论与应用, 2002, 19(5): 673-680) [11] Wang X F, Sun X Q, Feng Y J. Regional Mapping Model of Feedforword Neural Network for Pattern Recognition. Journal of Harbin Institute of Technology, 2000, 32(5):73-74 (in Chinese) (王雪峰,孙学全,冯英浚.用于模式识别的前馈式神经网络区域映射模型.哈尔滨工业大学学报, 2000, 32(5): 73-74) [12] Han M, Xi J H. Radial Basis Perceptron Networks and Its Application for Pattern Recognition. In: Proc of the International Joint Conference on Neural Networks. Honolulu, USA, 2002, 669-674
2006, Vol. 19 
