Improvements to the SMO algorithm for SVM regression
Improvements to the SMO algorithm for SVM regression
复制标题
DOI：
10.1109/72.870050
复制
发表时间：
2000-09-01
影响因子：
--
通讯作者：
Murthy, KRK
中科院分区：
文献类型：
--
作者：
Shevade, SK;Keerthi, SS;Murthy, KRK
本文指出了Smola和Scholkopf的支持向量机（St回归）的顺序最小优化（SMO）算法效率低下的一个重要原因，这是由于使用单一的阈值。利用对偶问题的KKT条件的线索，采用两个阈值参数来导出用于回归的SMO的修改，这些修改后的算法在所尝试的数据集上的执行速度明显快于原始SMO。
This paper points out an important source of inefficiency in Smola and Scholkopf's sequential minimal optimization (SMO) algorithm for support vector machine (St regression that is caused by the use of a single threshold value. Using clues from the KKT conditions for the dual problem, two threshold parameters are employed to derive modifications of SMO for regression, These modified algorithms perform significantly faster than the original SMO on the datasets tried.