摘要： Mangasarian 最近提出的用于分类的无参数鲁棒线性规划支持向量机克服了标准支持向量机的一些缺点，而且模型简单，容易实现。该文讨论了这种新型支持向量机的线性规划问题的最小2-范数解，在此基础上给出了一个快速的牛顿算法。
关键词:
支持向量机；无参数鲁棒线性规划支持向量机；牛顿算法；分类
Abstract: Parameterless rubost linear programming support vector machine for classification, proposed by Mangasarian recently, avoids some shortcomings in standard SVM. Furthermore, its formulation is simple and easy to realize. This paper discusses the least 2-norm solution of the linear programming formulation of the new SVM, and proposes a fast Newton method for it
Key words:
Support vector machine; Parameterless robust linear programming support vector machine; Newton algorithm; Classification
宋 杰，唐焕文. 无参数鲁棒线性规划支持向量机分类的牛顿法[J]. 计算机工程, 2006, 32(5): 10-12.
SONG Jie, TANG Huanwen. Newton Method for Parameterless Robust Linear Programming Support Vector Machine Classification[J]. Computer Engineering, 2006, 32(5): 10-12.