Learning Optimally Sparse Support Vector Machines
Proceedings of the 30th International Conference on Machine Learning, PMLR 28(1):266-274, 2013.
We show how to train SVMs with an optimal guarantee on the number of support vectors (up to constants), and with sample complexity and training runtime bounds matching the best known for kernel SVM optimization (i.e. without any additional asymptotic cost beyond standard SVM training). Our method is simple to implement and works well in practice.