A Kind of Neural Network Based on Chebyshev Orthogonal Polynomial and Imitating Example

摘要：

According to the characteristic of the orthogonal polynomials and the structure of RBF network,this paper proposed a kind of new artificial neural network. This network uses chebyshev orthogonal polynomial at interval -1,1 as the radial basis function, has the basic advantages of Radial-Basis-Function network as well as avoiding the trouble of seeking the centre and variance of Radial-Basis-Function. Computer simulation results show that the new artificial neural network has the higher speed convergence of the learning process than the conventional BP algorithm.

关键词： artificial neural network Radial-Basis-Function chebyshev orthogonal polynomials

作者: Di Wang Lixin Zhang Xinfeng Wang

作者单位: Department of Basic Course, Beijing Union University, Beijing, 100101, China

会议类型: 国际会议

会议名称: The 5th International Congress on Mathematical Biology(第五届国际生物数学大会 ICMB 2011)

会议地点: 南京

会议语种:英文

页码: 475-479

在线出版日期: 2011-06-01（万方平台首次上网日期，不代表论文的发表时间）