MREKLM: A fast multiple empirical kernel learning machine
详细信息 查看全文
- 作者：Qi FanaAuthor Vitae ; Zhe Wanga ; c ; wangzhe@ecust.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Hongyuan ZhabAuthor Vitae ; Daqi Gaoa ; gaodaqi@ecust.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Multiple Kernel Learning ; Empirical Kernel Mapping ; Random projection ; Analytical optimization ; Classifier design ; Pattern recognition
- 刊名：Pattern Recognition
- 出版年：2017
- 出版时间：January 2017
- 年：2017
- 卷：61
- 期：Complete
- 页码：197-209
- 全文大小：1095 K
文摘
- •
This paper proposes a fast Multiple Random Empirical Kernel Learning Machine (MREKLM).
- •
MREKLM employs an alternative Random Empirical Kernel Mapping (REKM) to construct low-dimensional feature spaces.
- •
MREKLM is of much lower computational and memory burden.
- •
MREKLM extends the capability of MEKL to handle the large-scale problems.