Improving Localized Multiple Kernel Learning via Radius-Margin Bound
Joint Authors
Du, Yajun
Wang, Xiaoming
Huang, Zengxi
Source
Mathematical Problems in Engineering
Issue
Vol. 2017, Issue 2017 (31 Dec. 2017), pp.1-12, 12 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2017-01-09
Country of Publication
Egypt
No. of Pages
12
Main Subjects
Abstract EN
Localized multiple kernel learning (LMKL) is an effective method of multiple kernel learning (MKL).
It tries to learn the optimal kernel from a set of predefined basic kernels by directly using the maximum margin principle, which is embodied in support vector machine (SVM).
However, LMKL does not consider the radius of minimum enclosing ball (MEB) which actually impacts the error bound of SVM as well as the separating margin.
In the paper, we propose an improved version of LMKL, which is named ILMKL.
The proposed method explicitly takes into consideration both the margin and the radius and so achieves better performance over its counterpart.
Moreover, the proposed method can automatically tune the regularization parameter when learning the optimal kernel.
Consequently, it avoids using the time-consuming cross-validation process to choose the parameter.
Comprehensive experiments are conducted and the results well demonstrate the effectiveness and efficiency of the proposed method.
American Psychological Association (APA)
Wang, Xiaoming& Huang, Zengxi& Du, Yajun. 2017. Improving Localized Multiple Kernel Learning via Radius-Margin Bound. Mathematical Problems in Engineering،Vol. 2017, no. 2017, pp.1-12.
https://search.emarefa.net/detail/BIM-1190442
Modern Language Association (MLA)
Wang, Xiaoming…[et al.]. Improving Localized Multiple Kernel Learning via Radius-Margin Bound. Mathematical Problems in Engineering No. 2017 (2017), pp.1-12.
https://search.emarefa.net/detail/BIM-1190442
American Medical Association (AMA)
Wang, Xiaoming& Huang, Zengxi& Du, Yajun. Improving Localized Multiple Kernel Learning via Radius-Margin Bound. Mathematical Problems in Engineering. 2017. Vol. 2017, no. 2017, pp.1-12.
https://search.emarefa.net/detail/BIM-1190442
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-1190442