Approximation Incremental Training Algorithm Based on a Changeable Training Set

Approximation Incremental Training Algorithm Based on a Changeable Training Set
复制标题

基于可变训练集的近似增量训练算法

DOI:
--
复制
发表时间:
2011-09
期刊:
International Journal of Computational and Mathematical Sciences
影响因子:
--
通讯作者:
Yong-lin Lei
Yong-lin Lei
中科院分区:
其他
文献类型:
--
作者:
Yi-fan Zhu;Wei Zhang;Xuan Zhou;Qun Li;Yong-lin Lei

文献摘要

参考文献

相似文献

增量学习的快速训练算法和精确求解过程旨在提高支持向量回归机的训练效率,但它们存在一些不足,即前者在训练集变化时不收敛,后者在大数据集时效率低下。针对这一问题,提出了一种新的可变训练集的训练算法--近似增量训练算法(AITA)。本文从理论上分析了AITA算法不收敛的原因,并讨论了AITA算法的实现,最后论证了AITA算法在精度和效率上的优势。关键词-支持向量回归,增量学习,可变训练集,快速训练算法,精确求解过程
The quick training algorithms and accurate solution procedure for incremental learning aim at improving the efficiency of training of SVR, whereas there are some disadvantages for them, i.e. the nonconvergence of the formers for changeable training set and the inefficiency of the latter for a massive dataset. In order to handle the problems, a new training algorithm for a changeable training set, named Approximation Incremental Training Algorithm (AITA), was proposed. This paper explored the reason of nonconvergence theoretically and discussed the realization of AITA, and finally demonstrated the benefits of AITA both on precision and efficiency. Keywords—support vector regression, incremental learning, changeable training set, quick training algorithm, accurate solution procedure
DOI: 10.1016/s0925-2312(02)00573-8
发表时间: 2003
期刊: Neurocomputing
影响因子: 6
作者:
Junbin Gao;S. Gunn;C. Harris
通讯作者: Junbin Gao;S. Gunn;C. Harris
DOI: --
发表时间: 2010
期刊: Application Research of Computers
影响因子: --
作者:
Wang Sheng-cai
通讯作者: Wang Sheng-cai
DOI: 10.5555/299094
发表时间: 1999-02
期刊: --
影响因子: --
作者:
B. Scholkopf;C. Burges;Alex Smola
通讯作者: B. Scholkopf;C. Burges;Alex Smola
DOI: 10.2514/6.2007-1912
发表时间: 2007-08
期刊: --
影响因子: --
作者:
Divija Odapally
通讯作者: Divija Odapally
DOI: --
发表时间: 2011-09
期刊: --
影响因子: --
作者:
유정완
通讯作者: 유정완