Tuning Support Vector Machines for Minimax and Neyman-Pearson Classification
Tuning Support Vector Machines for Minimax and Neyman-Pearson Classification
复制标题
DOI：
10.1109/tpami.2010.29
复制
发表时间：
2008-08
影响因子：
23.6
通讯作者：
M. Davenport;Richard Baraniuk;C. Scott
中科院分区：
文献类型：
--
作者：
M. Davenport;Richard Baraniuk;C. Scott
This paper studies the training of support vector machine (SVM) classifiers with respect to the minimax and Neyman-Pearson criteria. In principle, these criteria can be optimized in a straightforward way using a cost-sensitive SVM. In practice, however, because these criteria require especially accurate error estimation, standard techniques for tuning SVM parameters, such as cross-validation, can lead to poor classifier performance. To address this issue, we first prove that the usual cost-sensitive SVM, here called the 2C-SVM, is equivalent to another formulation called the 2\nu-SVM. We then exploit a characterization of the 2\nu-SVM parameter space to develop a simple yet powerful approach to error estimation based on smoothing. In an extensive experimental study, we demonstrate that smoothing significantly improves the accuracy of cross-validation error estimates, leading to dramatic performance gains. Furthermore, we propose coordinate descent strategies that offer significant gains in computational efficiency, with little to no loss in performance.
