Analysis of Generalization for a New Kind of Neural Network

摘要：

　　To describe the performance of a new kind of approximation algorithm using B-Spline weight functions on feedforward neural networks,the analysis of generalization is proposed in this paper.The neural networks architecture is very simple and the number of weight functions is independent of the number of patterns.Three important theorems are proved,which mean that,by increasing the density of the knots,the upper bound of the networks error can be made to approach zero.The results show that the new algorithm has good property of generalization and high learning speed.

关键词： artificial intelligence neural networks generalization weight functions B-spline functions

作者: Daiyuan Zhang

作者单位: College of Computer,Nanjing University of Posts and Telecommunications,Nanjing,China

会议类型: 国际会议

会议名称: 2013 2nd international Conference on Opto-Electronics Engineering and Materials Eesearch(2013第二届光电工程与材料研究国际会议)(OEMR2013)

会议地点: 郑州

会议语种:英文

页码: 911-915

在线出版日期: 2013-10-19（万方平台首次上网日期，不代表论文的发表时间）