Distance metric learning with penalized linear discriminant analysis

CORC > 沈阳自动化研究所 > 中国科学院沈阳自动化研究所 > 机器人学研究室

	Distance metric learning with penalized linear discriminant analysis
	Chen Y(陈洋); Zhao XG(赵新刚); Han JD(韩建达)
	2010
会议名称	2010 1st IEEE International Conference on Progress in Informatics and Computing, PIC 2010
会议日期	December 10-12, 2010
会议地点	Shanghai, China
关键词	Fisher information matrix Information science Learning algorithms Transfer matrix method
页码	170-174
中文摘要	Linear discriminant analysis has gained extensive applications in supervised classification and dimension reduction. In LDA formulation, original patterns with high dimension can be projected to lower dimension through a transfer matrix which is fundamental to clustering, nearest neighbor searches, and others. The transfer matrix is usually viewed as a distance metric. However, the classification accuracy under the LDA metric is neither optimal nor suboptimal because physical datasets often appear multimodal distribution. This paper proposes a penalized scheme for LDA to improve the classification rate by using the information of misclassified samples. This method is evaluated to be robust and effective by a great number of datasets from the machine learning repository. ©2010 IEEE.
收录类别	EI
产权排序	1
会议主办者	IEEE Beijing Section; Shanghai Jiao Tong University; University of Texas at Dallas (UTD); Osaka University
会议录	Proceedings of the 2010 IEEE International Conference on Progress in Informatics and Computing, PIC 2010
会议录出版者	IEEE Computer Society
会议录出版地	Piscataway, NJ
语种	英语
ISBN号	978-1-4244-6786-0
内容类型	会议论文
源URL	[http://ir.sia.cn/handle/173321/8672]
专题	沈阳自动化研究所_机器人学研究室
推荐引用方式 GB/T 7714	Chen Y,Zhao XG,Han JD. Distance metric learning with penalized linear discriminant analysis[C]. 见:2010 1st IEEE International Conference on Progress in Informatics and Computing, PIC 2010. Shanghai, China. December 10-12, 2010.