Global Convergence of SMO Algorithm for Support Vector Regression
Global Convergence of SMO Algorithm for Support Vector Regression
复制标题
DOI：
10.1109/tnn.2007.915116
复制
发表时间：
2008-06
影响因子：
--
通讯作者：
Norikazu Takahashi;Jun Guo;T. Nishi
中科院分区：
文献类型：
--
作者：
Norikazu Takahashi;Jun Guo;T. Nishi
本文研究了支持向量回归(SVR)的序贯最小优化(SMO)算法的全局收敛性。给定 l 个训练样本，SVR 被表述为具有 l 对变量的凸二次规划 (QP) 问题。我们证明，如果每一步选择两对违反最优条件的变量进行更新，并以某种方式解决子问题，那么SMO算法在找到最优解后总是在有限迭代次数内停止。此外，还介绍了 SMO 算法的有效实现技术，并与其他 SMO 算法进行了实验比较。
Global convergence of the sequential minimal optimization (SMO) algorithm for support vector regression (SVR) is studied in this paper. Given l training samples, SVR is formulated as a convex quadratic programming (QP) problem with l pairs of variables. We prove that if two pairs of variables violating the optimality condition are chosen for update in each step and subproblems are solved in a certain way, then the SMO algorithm always stops within a finite number of iterations after finding an optimal solution. Also, efficient implementation techniques for the SMO algorithm are presented and compared experimentally with other SMO algorithms.