三种训练光滑支持向量分类器方法的比较  

Comparison of three methods for training smooth support vector classification

作　　者：涂文根[1] 熊金志[1] 袁华强[1] 

机构地区：[1]东莞理工学院工程技术研究院,广东东莞523808

出　　处：《计算机工程与应用》2011年第3期190-195,共6页Computer Engineering and Applications

基　　金：国家自然科学基金No.60573029;No.60773050;广东省自然科学基金No.9151170003000017~~

摘　　要：光滑支持向量分类机(SSVC)是支持向量分类机(SVC)的快速求解模型,本质上是求解数学规划中具有光滑性和强凸性的无约束最优化问题。BFGS-Armijo和Newton-Armijo算法被用来训练SSVC,相比而言后者拥有更快的训练速度;牛顿-预优共轭梯度法(Newton-PCG)适用于求解无约束的最优化问题,理论上快于一般的Newton类算法。使用Newton-Armijo、BFGS-Armijo和Newton-PCG三种算法来训练光滑支持向量分类机,根据数值实验结果进行分析比较,证明了Newton-PCG算法有更优的效果。Smooth Support Vector Classification（SSVC） originally is unconstrained mathematical programming with convex and smooth,and is a method for solving Support Vector Classification（SVC） quickly.The BFGS-Armijo and Newton-Armijo algorithms have been used to train SSVC,and the latter has faster speed.Newton-PCG algorithm is just enough method for unconstrained problem which has better speed than Newton in theory.On the numerical experimentation of using BFGS-Armijo,Newton-Armijo and Newton-PCG to train SSVC,this paper gives analysis and comparison among the three algorithms,and proves that Newton-PCG has the best result.

关 键 词：模式识别 光滑支持向量机 分类 Newton-PCG算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]