Decomposition Methods for Linear Support Vector Machines
Authors:
- 1 August 2004
- journal article
- Published by MIT Press in Neural Computation
- Vol. 16 (8) , 1689-1704
Abstract
In this letter, we show that decomposition methods with alpha seeding are extremely useful for solving a sequence of linear support vector machines (SVMs) with more data than attributes. This strategy is motivated by Keerthi and Lin (2003), who proved that for an SVM with data not linearly separable, after C is large enough, the dual solutions have the same free and bounded components. We explain why a direct use of decomposition methods for linear SVMs is sometimes very slow and then analyze why alpha seeding is much more effective for linear than nonlinear SVMs. We also conduct comparisons with other methods that are efficient for linear SVMs and demonstrate the effectiveness of alpha seeding techniques in model selection.Keywords
This publication has 8 references indexed in Scilit:
- A study on reduced support vector machinesIEEE Transactions on Neural Networks, 2003
- Asymptotic Behaviors of Support Vector Machines with Gaussian KernelNeural Computation, 2003
- A comparison of methods for multiclass support vector machinesIEEE Transactions on Neural Networks, 2002
- A Simple Decomposition Method for Support Vector MachinesMachine Learning, 2002
- 10.1162/15324430152748218Applied Physics Letters, 2000
- 10.1162/15324430260185619Applied Physics Letters, 2000
- A fast iterative nearest point algorithm for support vector machine classifier designIEEE Transactions on Neural Networks, 2000
- Support-vector networksMachine Learning, 1995