A training algorithm for optimal margin classifiers
COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
The nature of statistical learning theory
The nature of statistical learning theory
Machine Learning
Combining labeled and unlabeled data with co-training
COLT' 98 Proceedings of the eleventh annual conference on Computational learning theory
Making large-scale support vector machine learning practical
Advances in kernel methods
Feature Selection via Concave Minimization and Support Vector Machines
ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
Training Support Vector Machines: an Application to Face Detection
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Dimensionality reduction via sparse support vector machines
The Journal of Machine Learning Research
Estimation of Dependences Based on Empirical Data: Empirical Inference Science (Information Science and Statistics)
Kernel Based Algorithms for Mining Huge Data Sets: Supervised, Semi-supervised, and Unsupervised Learning (Studies in Computational Intelligence)
Semi-Supervised Learning
| Hi-index | 0.00 |
In the process of human learning, teachers always play an important role. However, for most of the existing machine learning method, the role of teachers is seldom considered. Recently, Vapnik introduced an advanced learning paradigm called Learning Using Privileged Information(LUPI) to include the elements of human teaching in machine learning. Through theoretical analysis and numerical experiments, the superiority of LUPI over the classical learning paradigm has received preliminary proof. In this paper, on the basis of existing work for LUPI, we introduce the privileged information into the modeling of L-1 support vector machine(SVM). Compared with the existing research of LUPI with L-2 SVM, the new method has the advantage of spending less time on tuning model parameters and the additional benefits of performing feature selection in the training process. Experiments on the digit recognition problem validate the effectiveness of our method.