支持向量机回归的序列最小优化算法改进

Improvements to the SMO algorithm for SVM regression.

作者信息

Shevade S K, Keerthi S S, Bhattacharyya C, Murthy K K

机构信息

Department of Computer Science and Automation, Indian Institute of Science, Bangalore 560012, India.

出版信息

IEEE Trans Neural Netw. 2000;11(5):1188-93. doi: 10.1109/72.870050.

DOI:10.1109/72.870050

PMID:18249845

Abstract

This paper points out an important source of inefficiency in Smola and Schölkopf's sequential minimal optimization (SMO) algorithm for support vector machine (SVM) 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.

摘要

本文指出了支持向量机（SVM）回归的斯莫拉和施尔科普夫序列最小优化（SMO）算法中一个重要的低效率来源，该来源是由使用单个阈值所导致的。利用对偶问题的KKT条件中的线索，采用两个阈值参数来推导用于回归的SMO的改进算法。在尝试的数据集上，这些改进后的算法比原始的SMO运行速度明显更快。