###
Journal of Software:2012.23(2):279-288

基于最大相关熵准则的鲁棒半监督学习算法
杨南海,黄明明,赫然,王秀坤
(大连理工大学 计算机科学与技术学院,辽宁 大连 116024;大连理工大学 计算机科学与技术学院,辽宁 大连 116024; 模式识别国家重点实验室(中国科学院 自动化研究所),北京 100190)
Robust Semi-Supervised Learning Algorithm Based on Maximum Correntropy Criterion
YANG Nan-Hai,HUANG Ming-Ming,HE Ran,WANG Xiu-Kun
(School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, China;School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, China; National Laboratory of Pattern Recognition (Institute of Automation, The Chinese Academy of Sciences), Beijing 100190, China)
Abstract
Chart / table
Reference
Similar Articles
Article :Browse 4221   Download 5365
Received:May 04, 2010    Revised:October 11, 2010
> 中文摘要: 分析了噪声对半监督学习Gaussian-Laplacian 正则化(Gaussian-Laplacian regularized,简称GLR)框架的影响,针对最小二乘准则对噪声敏感的特点,结合信息论的最大相关熵准则(maximum correntropy criterion,简称MCC), 提出了一种基于最大相关熵准则的鲁棒半监督学习算法(简称GLR-MCC),并证明了算法的收敛性.半二次优化技术被用来求解相关熵目标函数.在每次迭代中,复杂的信息论优化问题被简化为标准的半监督学习问题.典型机器学习数据集上的仿真实验结果表明,在标签噪声和遮挡噪声的情况下,该算法能够有效地提高半监督学习算法性能.
Abstract:This paper analyzes the problem of sensitivity to noise in the mean square criterion of Gaussian- Laplacian regularized (GLR) algorithm. A robust semi-supervised learning algorithm based on maximum correntropy criterion (MCC), called GLR-MCC, is proposed to improve the robustness of GLR along with its convergence analysis. The half quadratic optimization technique is used to simplify the correntropy optimization problem to a standard semi-supervised problem in each iteration. Experimental results on typical machine learning data sets show that the proposed GLR-MCC can effectively improve the robustness of mislabeling noise and occlusion as compared with related semi-supervised learning algorithms.
文章编号:     中图分类号:    文献标志码:
基金项目:国家自然科学基金(60873054, 50909012); 国家教育部高等学校博士学科点专项科研基金(20100041120009) 国家自然科学基金(60873054, 50909012); 国家教育部高等学校博士学科点专项科研基金(20100041120009)
Foundation items:
Reference text:

杨南海,黄明明,赫然,王秀坤.基于最大相关熵准则的鲁棒半监督学习算法.软件学报,2012,23(2):279-288

YANG Nan-Hai,HUANG Ming-Ming,HE Ran,WANG Xiu-Kun.Robust Semi-Supervised Learning Algorithm Based on Maximum Correntropy Criterion.Journal of Software,2012,23(2):279-288