Fast Algorithms for Linear and Kernel SVM+
Authors:
- Wen LiWLWen Li
- Dengxin DaiDDDengxin Dai
- Mingkui TanMTMingkui Tan
- Dong XuDXDong Xu
- Luc Van GoolLGLuc Van Gool
- 1 June 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2258-2266
Abstract
The SVM+ approach has shown excellent performance in visual recognition tasks for exploiting privileged information in the training data. In this paper, we propose two efficient algorithms for solving the linear and kernel SVM+, respectively. For linear SVM+, we absorb the bias term into the weight vector, and formulate a new optimization problem with simpler constraints in the dual form. Then, we develop an efficient dual coordinate descent algorithm to solve the new optimization problem. For kernel SVM+, we further apply the l2-loss, which leads to a simpler optimization problem in the dual form with only half of dual variables when compared with the dual form of the original SVM+ method. More interestingly, we show that our new dual problem can be efficiently solved by using the SMO algorithm of the one-class SVM problem. Comprehensive experiments on three datasets clearly demonstrate that our proposed algorithms achieve significant speed-up than the state-of-the-art solvers for linear and kernel SVM+.Keywords
This publication has 15 references indexed in Scilit:
- ImageNet classification with deep convolutional neural networksCommunications of the ACM, 2017
- Is image super-resolution helpful for other vision tasks?2016
- Metric imitation by manifold transfer for efficient vision applications2015
- Caffe2014
- Recognizing RGB Images by Learning from RGB-D Data2014
- Learning to Rank Using Privileged Information2013
- Incorporating Privileged Information Through Metric LearningIEEE Transactions on Neural Networks and Learning Systems, 2013
- NUS-WIDE2009
- A new learning paradigm: Learning using privileged informationNeural Networks, 2009
- A dual coordinate descent method for large-scale linear SVM2008