Minimal Complexity Support Vector Machines
Minimal Complexity Support Vector Machines
复制标题
最小复杂度支持向量机
DOI：
10.1007/978-3-030-58309-5_7
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Shigeo Abe
中科院分区：
文献类型：
--
作者：
Shigeo Abe;Shigeo Abe
最小复杂性机器（MCMs）最小化VC（Vapnik-Chervonenkis）维，以获得高的泛化能力。然而，由于目标函数中不包含正则化项，因此解并不唯一。针对这一问题，本文提出了将MCM与标准支持向量机（L1 SVM）相融合的方法。这是通过最小化L1 SVM中训练数据的决策函数上界来实现的。我们将该机器称为最小复杂度L1 SVM（ML 1 SVM）。我们比较了ML 1 SVM与其他类型的SVM，包括L1 SVM使用几个基准数据集，并表明ML 1 SVM的性能相当或优于L1 SVM。
Minimal complexity machines (MCMs) minimize the VC (Vapnik-Chervonenkis) dimension to obtain high generalization abilities. However, because the regularization term is not included in the objective function, the solution is not unique. In this paper, to solve this problem, we propose fusing the MCM and the standard support vector machine (L1 SVM). This is realized by minimizing the upper bound on the decision function for the training data in the L1 SVM. We call the machine Minimum complexity L1 SVM (ML1 SVM). We compare the ML1 SVM with other types of SVMs including the L1 SVM using several benchmark data sets and show that the ML1 SVM performs comparable to or better than the L1 SVM.