Multiple kernel active learning for facial expression analysis

Authors:
Siyao Fu;Xinkai Kuai;Guosheng Yang
Affiliations:
School of Information and Engineering, the Central University of Nationalities, Beijing, China;School of Information and Engineering, the Central University of Nationalities, Beijing, China;School of Information and Engineering, the Central University of Nationalities, Beijing, China
Venue:
ISNN'11 Proceedings of the 8th international conference on Advances in neural networks - Volume Part II
Year:
2011

Citing 7
Cited 0

Support vector machine active learning with applications to text classification

The Journal of Machine Learning Research
Learning the Kernel Matrix with Semidefinite Programming

The Journal of Machine Learning Research
Multiple kernel learning, conic duality, and the SMO algorithm

ICML '04 Proceedings of the twenty-first international conference on Machine learning
Large Scale Multiple Kernel Learning

The Journal of Machine Learning Research
More efficiency in multiple kernel learning

Proceedings of the 24th international conference on Machine learning
Image category learning and classification via optimal linear combination of multiple partially matching kernels

Soft Computing - A Fusion of Foundations, Methodologies and Applications - Special Issue on Pattern Recognition and Information Processing Using Neural Networks;Guest Editors: Fuchun Sun,Ying Tan,Cong Wang
Multiple kernel active learning for image classification

ICME'09 Proceedings of the 2009 IEEE international conference on Multimedia and Expo

Quantified Score

Hi-index	0.00

Visualization

Abstract

Multiple Kernel Learning (MKL) approaches aim at determine the optimal combination of similarity matrices (since each representation leads to a different similarity measure between images, thus, kernel functions) and the optimal classifier simultaneously. However, the combination of "passive" kernels learning scheme limits MKL's efficiency because side information is provided beforehand. A framework of Multiple Kernel Active Learning (MKAL) is presented in this paper, in which the most informative exemplars are efficiently selected by min-max algorithm, the margin ratio is used for querying next instance. We demonstrate our algorithm on facial expression categorization tasks, showing that the proposed method is accurate and more efficient than current approaches.