文章基本信息

标题：On the Relationship between LS-SVM, MSA, and LSA
本地全文：下载
作者：Yatong Zhou ; Taiyi Zhang ; Liejun Wang 等
期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2006
卷号：6
期号：11
页码：1-5
出版社：International Journal of Computer Science and Network Security
摘要：A common task in signal processing is to approximate adequately a signal. It is crucial to understand various methods to this task. In this paper we survey three different approximation methods: least squares support vector machine (LS-SVM), multiresolution signal approximation (MSA), and least squares approximation (LSA) to highlight their mathematical relationship. Based on the theoretical analysis, we prove that LS-SVM is identical to the LSA with the minimum norm solution, and MSA is identical to the LSA with the least squares solution. Therefore, both LS-SVM and MSA can be derived as specific instances of LSA.
关键词：Signal processing, support vector machine, multiresolution, approximation, least squares, relationship