An EnKF-based scheme to optimize hyper-parameters and features for SVM classifier
详细信息 查看全文
- 作者：Yingsheng Jia ; bAuthor Vitae ; Yushu ChenbAuthor Vitae ; Haohuan FuaAuthor Vitae ; Guangwen Yanga ; b ; ygw@mail.tsinghua.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：EnKF ; SVM ; Hyper-parameter optimization ; Feature weighting ; Feature selection
- 刊名：Pattern Recognition
- 出版年：2017
- 出版时间：February 2017
- 年：2017
- 卷：62
- 期：Complete
- 页码：202-213
- 全文大小：1087 K
文摘
- class="listitem" id="list_li0005">
- class="label">•
We adopt the EnKF based scheme to optimize hyper-parameters and features.
- class="label">•
We build the EnKF based framework for parameter optimization.
- class="label">•
Multiple ensemble and ensemble evolution are proposed to enhance effects.
- class="label">•
Householder method is proposed to optimize the EnKF analysis computation.
- class="label">•
The EnKF scheme outperforms three current Bayesian optimization methods.