支持向量机与神经网络的关系研究  

The Researches on the Relationship between Support Vector Machine and Neural Network

作　　者：沈正维[1] 李秋菊[2] 

机构地区：[1]辽宁师范大学数学学院,辽宁大连116029 [2]鞍山师范学院数学系,辽宁鞍山114007

出　　处：《生物数学学报》2006年第2期204-208,共5页Journal of Biomathematics

基　　金：辽宁省教育厅科学研究计划资助(2004C068)

摘　　要：支持向量机是一种基于统计学习理论的新颖的机器学习方法,由于其出色的学习性能,该技术已成为当前国际机器学习界的研究热点,该方法已经广泛用于解决分类和回归问题．本文将结构风险函数应用于径向基函数网络学习中,同时讨论了支持向量回归模型和径向基函数网络之间的关系．仿真实例表明所给算法提高了径向基函数网络的泛化性能．Support vector machines are a kind of novel machine learning methods, based on statistical learning theory, which have become the hotspot of machine learning because of their excellent learning performance, and the method of support vector machines has been developed for solving classification and regression problems. This paper applies the structure risk function to radial basis function （RBF） networks, and then the relationship between support vector regression model and RBF networks is discussed. Simulation experiments show that this algorithm improves generalization ability of RBF networks.

关 键 词：支持向量机 神经网络 回归 结构风险函数 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]