Robustness of learning algorithms using hinge loss with outlier indicators
Robustness of learning algorithms using hinge loss with outlier indicators
复制标题
DOI：
10.1016/j.neunet.2017.07.005
复制
发表时间：
2017-10
期刊：
影响因子：
--
通讯作者：
T. Kanamori;Shuhei Fujiwara;A. Takeda
中科院分区：
文献类型：
--
作者：
T. Kanamori;Shuhei Fujiwara;A. Takeda
我们提出了一个统一的配方的强大的学习方法分类和回归问题。在学习方法中，铰链损失与异常值指标一起使用，以检测观察到的数据中的异常值。为了分析鲁棒性，我们在离群值比率不一定很小的情况下评估了学习方法的崩溃点。虽然最小化铰链损失与离群指标是一个非凸优化问题，我们证明了我们的学习算法的任何局部最优解具有鲁棒性。数值实验证实了理论研究结果。
We propose a unified formulation of robust learning methods for classification and regression problems. In the learning methods, the hinge loss is used with outlier indicators in order to detect outliers in the observed data. To analyze the robustness property, we evaluate the breakdown point of the learning methods in the situation that the outlier ratio is not necessarily small. Although minimization of the hinge loss with outlier indicators is a non-convex optimization problem, we prove that any local optimal solution of our learning algorithms has the robustness property. The theoretical findings are confirmed in numerical experiments.