Multi-class kernel margin maximization for kernel learning

Multi-class kernel margin maximization for kernel learning
复制标题

用于内核学习的多类内核边际最大化

DOI:
10.1016/j.neucom.2016.05.073
复制
发表时间:
2016-09
期刊:
影响因子:
6
通讯作者:
Song Zhan
Song Zhan
中科院分区:
计算机科学2区
文献类型:
--
作者:
Zhao Yan-Guo;Li Miaomiao;Chung Ronald;Song Zhan

文献摘要

参考文献

被引文献

相似文献

Two-stage multiple kernel learning (MKL) algorithms have been intensively studied due to its high efficiency and effectiveness. Pioneering work on this regard attempts to optimize the combination coefficients by maximizing the multi-class margin of a kernel, while obtaining unsatisfying performance. In this paper, we attribute this poor performance to the way in calculating the multi-class margin of a kernel. In specific, we argue that for each sample only thek-nearest neighbors, while not all samples with the same label, should be selected for calculating the margin. After that, we also develop another sparse variant which is able to automatically identify the nearest neighbors and the corresponding weights of each sample. Extensive experimental results on ten UCI data sets and six MKL benchmark data sets demonstrate the effectiveness and efficiency of the proposed algorithms.
DOI: 10.1109/tsmcb.2012.2212243
发表时间: 2013-04-01
影响因子: 11.8
作者:
Liu, Xinwang;Wang, Lei;Zhang, Jian
通讯作者: Zhang, Jian
DOI: --
发表时间: 2012-06
期刊: --
影响因子: --
作者:
Abhishek Kumar;Alexandru Niculescu-Mizil;K. Kavukcuoglu;Hal Daumé
通讯作者: Abhishek Kumar;Alexandru Niculescu-Mizil;K. Kavukcuoglu;Hal Daumé
DOI: 10.1609/aaai.v28i1.8983
发表时间: 2014-06
期刊: --
影响因子: --
作者:
Xinwang Liu;Lei Wang;Jian Zhang;Jianping Yin
通讯作者: Xinwang Liu;Lei Wang;Jian Zhang;Jianping Yin
DOI: --
发表时间: 2013-06
期刊: --
影响因子: --
作者:
Corinna Cortes;M. Mohri;Afshin Rostamizadeh
通讯作者: Corinna Cortes;M. Mohri;Afshin Rostamizadeh
DOI: --
发表时间: 2012-06
期刊: --
影响因子: --
作者:
M. Gönen
通讯作者: M. Gönen