文献类型：会议论文

英文题名：Ramp Loss Linear Programming Nonparallel Support Vector Machine

作者：Liu, Dalian[1,2];Chen, Dandan[3,4];Shi, Yong[1,4,5,6];Tian, Yingjie[4,5]

第一作者：Liu, Dalian

通讯作者：Liu, DL[1];Liu, DL[2]

机构：[1]Beijing Jiaotong Univ, Sch Comp & Informat Technol, Beijing 100044, Peoples R China;[2]Beijing Union Univ, Dept Basic Course Teaching, Beijing 100101, Peoples R China;[3]Univ Chinese Acad Sci, Coll Math Sci, Beijing 100049, Peoples R China;[4]Univ Chinese Acad Sci, Res Ctr Fictitious Econ & Data Sci, Beijing 100190, Peoples R China;[5]Univ Chinese Acad Sci, Key Lab Big Data Min & Knowledge Management, Beijing 100190, Peoples R China;[6]Univ Nebraska, Coll Informat Sci & Technol, Omaha, NE 68182 USA

第一机构：Beijing Jiaotong Univ, Sch Comp & Informat Technol, Beijing 100044, Peoples R China

通讯机构：[1]corresponding author), Beijing Jiaotong Univ, Sch Comp & Informat Technol, Beijing 100044, Peoples R China;[2]corresponding author), Beijing Union Univ, Dept Basic Course Teaching, Beijing 100101, Peoples R China.|[1141788]北京联合大学基础教学部;[11417]北京联合大学;

会议论文集：16th Annual International Conference on Computational Science (ICCS)

会议日期：JUN 06-08, 2016

会议地点：Univ Calif, San Diego Supercomputer Ctr, San Diego, CA

主办单位：Univ Calif, San Diego Supercomputer Ctr

语种：英文

外文关键词：support vector machine; nonparallel; CCCP; linear programming; ramp loss

摘要：Motivated by the fact that the l(1)-penalty is piecewise linear, we proposed a ramp loss linear programming nonparallel support vector machine (ramp-LPNPSVM.), in which the l(1)-penalty is applied for the RNPSVM, for binary classification. Since the ramp loss has the piecewise linearity as well, ramp-LPNPSVM. is a piecewise linear minimization problem and a local minimum can be effectively found by the Concave Convex Procedure and experimental results on benchmark datasets confirm the effectiveness of the proposed algorithm. Moreover, the l(1)-penalty can enhance the sparsity.