Adaptive margin slack minimization in RKHS for classification
Paper in proceeding, 2016
In this paper, we design a novel regularized empirical risk minimization technique for classification called Adaptive Margin Slack Minimization (AMSM). The proposed method is based on minimizing a regularized upper bound of the misclassification error. Compared to the cost function of the classical L2-SVM, AMSM can be interpreted as minimizing a tighter bound with some additional flexibilities regarding the choice of marginal hyperplane. A hyperparameter-free adaptive algorithm is presented for finding a solution to the proposed risk function. Numerical results shows that AMSM outperforms L2-SVM on the tested standard datasets.
Structural Risk Minimization
L2-SVM
Adaptive Margin
Reproducing Kernel Hilbert Space
Author
Yinan Yu
Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering
Konstantinos I. Diamantaras
TEI of Thessaloniki
Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering
Princeton University
Tomas McKelvey
Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering
S. Y. Kung
TEI of Thessaloniki
Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering
Princeton University
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
15206149 (ISSN)
Vol. 2016-May 2319-2323978-1-4799-9988-0 (ISBN)
Subject Categories
Signal Processing
DOI
10.1109/ICASSP.2016.7472091
ISBN
978-1-4799-9988-0