双正则化参数的L_2-SVM参数选择

Parameter optimization of L_2-SVM with two regularization parameters

寻找支持向量机(SVM)的最优参数是支持向量机研究领域的热点之一。2范数软间隔SVM(L2-SVM)将样本转化成线性可分,在原始单正则化参数L2-SVM的基础上,提出双正则化参数的L2-SVM,获得它的对偶形式,从而确定了最优化的目标函数。然后结合梯度法,提出了一种新的支持向量机参数选择的新方法(Doupenalty-Gradient)。实验使用了10个基准数据集,结果表明,Doupenalty-Gradient方法是可行且有效的。对于实验所用的样本,极大地改善了分类精度。

Searching the optimal parameters is one of the most important area of SVM and often named as parameter opti-mization or parameter selection. The L2-SVM can convert the samples into linearly separable problem. Based on the per-formance, this paper proposes the L2-SVM with two regularization parameters, and the dual formulation of L2-SVM with two regularization parameters is deduced. Combining the objective function established on minimizing the VC dimension and the gradient method, a new algorithm called Doupenalty-Gradient is present. Ten benchmark datasets are used in the experiments, and the classifying accuracy is improved obviously. The experimental results show the wonderful property and the feasibility of Doupenalty-Gradient.