DocumentCode :
457241
Title :
Weakly Supervised Learning on Pre-image Problem in Kernel Methods
Author :
Zheng, Wei-Shi ; Lai, Jian-Huang ; Yuen, Pong C.
Author_Institution :
Dept. of Math., Sun Yat-sen Univ., Guangzhou
Volume :
2
fYear :
0
fDate :
0-0 0
Firstpage :
711
Lastpage :
715
Abstract :
This paper presents a novel alternative approach, namely weakly supervised learning (WSL), to learn the pre-image of a feature vector in the feature space induced by a kernel. It is known that the exact pre-image may typically seldom exist, since the input space and the feature space are not isomorphic in general, and an approximate solution is required in past. The proposed WSL, however, would find an appropriate rather than only a purely approximate solution. WSL is able to involve some weakly supervised prior knowledge into the study of pre-image. The prior knowledge is weak and no class label of the sample is required, providing only information of positive class and negative class which should properly depend on applications. The proposed algorithm is demonstrated on kernel principal component analysis (KPCA) with application to illumination normalization and image denoising on faces. Evaluations of the performance of the proposed algorithm show notable improvement as comparing with some well-known existing approaches
Keywords :
feature extraction; image denoising; learning (artificial intelligence); principal component analysis; feature space; feature vector; illumination normalization; image denoising; kernel methods; kernel principal component analysis; preimage problem; weakly supervised learning; Image denoising; Kernel; Least squares approximation; Lighting; Machine learning; Mathematics; Noise reduction; Pattern recognition; Sun; Supervised learning;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 2006. ICPR 2006. 18th International Conference on
Conference_Location :
Hong Kong
ISSN :
1051-4651
Print_ISBN :
0-7695-2521-0
Type :
conf
DOI :
10.1109/ICPR.2006.1187
Filename :
1699304
Link To Document :
بازگشت