迭代重加权最小二乘支持向量机快速算法研究

计算机科学 ›› 2010, Vol. 37 ›› Issue (8): 224-228.

迭代重加权最小二乘支持向量机快速算法研究

温雯,郝志峰,邵壮丰

(广东工业大学计算机学院广州510006);(华南理工大学计算机科学与工程学院广州510641);(中国电信广东互联网与增值业务运营中心广州510110)

出版日期:2018-12-01 发布日期:2018-12-01
基金资助:
本文受信息安全国家重点实验室开放课题基金(20090401)资助。

Study on the Fast Training Algorithm of Iteratively Re-weighted Least Squares Support Vector Machine

WEN Wen,HAO Zhi-feng,SHAO Zhuang-feng

Online:2018-12-01 Published:2018-12-01

摘要/Abstract

摘要： 迭代重加权(Iteratively Reweighted)方法是提高最小二乘支持向量机(LS-SVM)稳健性的重要手段，但由于涉及到多次加权和重复训练，该方法需要大量运算，无法广泛应用。通过数值推导，获得了求解迭代重加权最小二乘支持向量机(IRLS-SVM)的快速算法，大幅度减少了其运算复杂度。引入了3种经典的加权函数，并在多个仿真数据集和实际数据集上进行实验，证实了IRLS-SVM能获得相当稳健的学习结果，所提出的快速算法也确实能够大幅度减少训练时间。实验结果同时表明，在快速训练算法的框架下，3种不同的权重函数可能要求不同的训练时间。

关键词: 支持向量机，稳健性，异常样本，快速算法

Abstract: Iteratively reweighted method is an important approach to improve the robustness of least sctuares support vector machine(LS-SVM). However, the reweighting and retraining procedure demands a lot of computational time, which makes it impossible for practical applications. In this paper, the iteratively reweighted least squares support vector machine (IRLS-SVM) was studied. An improved training algorithm of IRLS-SVM was proposed. It is based on novel numerical method, and can effectively reduce the computational complexity of IRIS-SVM. Three different weight funclions were implemented in the IRLS-SVM. Experiments on simulated instances and real-world datasets demonstrate the validity of this algorithm. Meanwhile, the results reveal that different weight function may require different computational time for the fast training algorithm of IRLS-SVM.

Key words: Support vector machines,Robustness,Outliers,Fast algorithm

温雯,郝志峰,邵壮丰. 迭代重加权最小二乘支持向量机快速算法研究[J]. 计算机科学, 2010, 37(8): 224-228. https://doi.org/

WEN Wen,HAO Zhi-feng,SHAO Zhuang-feng. Study on the Fast Training Algorithm of Iteratively Re-weighted Least Squares Support Vector Machine[J]. Computer Science, 2010, 37(8): 224-228. https://doi.org/