摘要
针对隐空间支持向量机求解约束凸二次规划存在训练时间长、计算复杂的问题,提出粒子群优化的隐空间光滑支持向量机算法(PSO-HSSSVM)。该算法通过隐函数将训练样本映射至隐空间,且对隐函数没有正定性限制;在隐空间中以熵函数近似松弛向量的加函数,导出光滑可微的无约束凸二次规划;引入共轭梯度法求解光滑模型,并采用粒子群优化算法选取最优参数。数值实验表明,PSO-HSSSVM算法可拓宽光滑支持向量机的核函数范围,而其训练精度和训练时间与光滑支持向量机的相当;PSO-HSSSVM算法可将隐空间支持向量机的训练精度提高2.14%,而其训练时间仅为隐空间支持向量机的9.5%。
A hidden space smooth support vector machine with particle swarm optimization （PSO- HSSSVM） algorithm is proposed to solve problems of long training time and computing complex in using hidden space support vector machine （HSSVM） to solve constrained convex quadratic programs. The proposed algorithm transforms the input data to a hidden space using a hidden function, and has no any restriction on the positive definity of the hidden function. The entropy function is employed to approximate the plus function of the slack vector, and a smooth differentiable unconstrained convex quadratic program is derived. The conjugate gradient （CG） algorithm is used to solve the smooth model, and the particle swarm optimization （PSO） algorithm is used to give the optimal parameters. Experiments show that the PSO-HSSSVM enlarges the usable kernels of smooth support vector machine （SSVM）, and its accuracy and training time are similar to those of SSVM; The PSO-HSSSVM improves the accuracy of HSSVM by 2.14%, and the training time is only 9.5% that of HSSVM.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2013年第12期38-42,共5页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(61373174)
陕西省教育厅专项科研计划资助项目(2010JK773)
西安石油大学博士专项科研基金资助项目(YS29030903)
关键词
隐空间
支持向量机
熵函数
粒子群优化
共轭梯度法
hidden space
support vector machine~ entropy function
particle swarm optimiza-tion~ conjugate graduate
作者简介
梁锦锦（1983-）,女,博士,讲师.
