UM
Information-theoretic atomic representation for robust pattern classification
Wang Y.2; Tang Y.Y.2; Li L.3; Wang P.1
2017-04-13
Source PublicationProceedings - International Conference on Pattern Recognition
Pages3685-3690
AbstractRepresentation-based classifiers (RCs) including sparse RC (SRC) have attracted intensive interest in pattern recognition in recent years. In our previous work, we have proposed a general framework called atomic representation-based classifier (ARC) including many popular RCs as special cases. Despite the empirical success, ARC and conventional RCs utilize the mean square error (MSE) criterion and assign the same weights to all entries of the test data, including both severely corrupted and clean ones. This makes ARC sensitive to the entries with large noise and outliers. In this work, we propose an information-theoretic ARC (ITARC) framework to alleviate such limitation of ARC. Using ITARC as a general platform, we develop three novel representation-based classifiers. The experiments on public real-world datasets demonstrate the efficacy of ITARC for robust pattern recognition.
DOI10.1109/ICPR.2016.7900207
URLView the original
Language英語
Fulltext Access
Citation statistics
Document TypeConference paper
CollectionUniversity of Macau
Affiliation1.College of Computer and Information Science
2.Universidade de Macau
3.Hubei University
Recommended Citation
GB/T 7714
Wang Y.,Tang Y.Y.,Li L.,et al. Information-theoretic atomic representation for robust pattern classification[C],2017:3685-3690.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wang Y.]'s Articles
[Tang Y.Y.]'s Articles
[Li L.]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wang Y.]'s Articles
[Tang Y.Y.]'s Articles
[Li L.]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wang Y.]'s Articles
[Tang Y.Y.]'s Articles
[Li L.]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.