基于改进RM界的二次损失函数支持向量机模式选择

Model Selection for 2-norm Support Vector Machine Based on Improved RM Bound

半径-间隔界中最小包含球半径R的计算需要求解二次规划问题,增加了算法的计算量。为提高计算效率,提出一种基于改进RM界的二次损失函数支持向量机模式选择。用所有训练样本的最大距离D逼近半径R,用D替换R构成新的RM界,然后基于改进的RM界对二次损失函数支持向量机(L2-SVM)进行模式选择,并用梯度下降法调节最优参数。对算法的分类精度和计算效率进行仿真实验讨论,结果表明,与基于RM界的模式选择相比,虽然该算法的分类精度没有明显改变,但其计算效率至少提高1倍。

Calculating the radius of radius-margin（ RM） bound by solving the quadratic programming adds the computational overload. In order to solve this problem,we construct a new RM bound which approximates the radius by using the maximum pairwise distance over all points. Then based on new RM bound,the model selection of 2-norm SVM（ L2-SVM） was conducted,and automatically adjusted parameters by employing the gradient descent algorithm. Finally,the classification accuracy and computational efficiency of the algorithm were discussed through simulation experiments. The experimental results show that the classification accuracy of the proposed algorithm is not significantly changed compared with the model selection based on RM bound,but the computational efficiency is improved at least one fold.