跳到主要导航 跳到搜索 跳到主要内容

A novel manifold regularized online semi-supervised learning algorithm

  • Shuguang Ding
  • , Xuanyang Xi
  • , Zhiyong Liu
  • , Hong Qiao
  • , Bo Zhang
  • CAS - Institute of Applied Mathematics
  • CAS - Institute of Automation
  • CAS Center for Excellence in Brain Science and Intelligence Technology

科研成果: 书/报告/会议事项章节会议稿件同行评审

1 引用 (Scopus)

摘要

In this paper, we propose a novel manifold regularized online semi-supervised learning (OS2L) model in an Reproducing Kernel Hilbert Space (RK-HS). The proposed algorithm, named Model-Based Online Manifold Regularization (MOMR), is derived by solving a constrained optimization problem, which is different from the stochastic gradient algorithm used for solving the online version of the primal problem of Laplacian support vector machine (LapSVM). Taking advantage of the convex property of the proposed model, an exact solution can be obtained iteratively by solving its Lagrange dual problem. Furthermore, a buffering strategy is introduced to improve the computational efficiency of the algorithm. Finally, the proposed algorithm is experimentally shown to have a comparable performance to the standard batch manifold regularization algorithm.

源语言英语
主期刊名Neural Information Processing - 23rd International Conference, ICONIP 2016, Proceedings
编辑Kenji Doya, Kazushi Ikeda, Minho Lee, Akira Hirose, Seiichi Ozawa, Derong Liu
出版商Springer Verlag
597-605
页数9
ISBN(印刷版)9783319466866
DOI
出版状态已出版 - 2016
已对外发布
活动23rd International Conference on Neural Information Processing, ICONIP 2016 - Kyoto, 日本
期限: 16 10月 201621 10月 2016

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
9947 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议23rd International Conference on Neural Information Processing, ICONIP 2016
国家/地区日本
Kyoto
时期16/10/1621/10/16

学术指纹

探究 'A novel manifold regularized online semi-supervised learning algorithm' 的科研主题。它们共同构成独一无二的指纹。

引用此