期刊
APPLIED MATHEMATICAL MODELLING
卷 82, 期 -, 页码 235-251出版社
ELSEVIER SCIENCE INC
DOI: 10.1016/j.apm.2020.01.053
关键词
Support vector regression; Nonconvex loss; Robust regression; Concave-convex programming
资金
- National Natural Science Foundation of China [11871183, 61866010, 11926349, 61703370, 61603338]
- Zhejiang Provincial Natural Science Foundation of China [LY18G010018, LQ19D010001, LQ17F030003]
In this paper, we propose a robust support vector regression with a novel generic non-convex quadratic epsilon-insensitive loss function. The proposed method is robust to outliers or noise since it can adaptively control the loss value and decrease the negative influence of outliers or noise on the decision function by adjusting the elastic interval parameter and adaptive robustification parameter. Given the nature of the nonconvexity of the optimization problem, a concave-convex programming procedure is employed to solve the proposed problem. Experimental results on two artificial data sets and three real-world data sets indicate that the proposed method outperforms support vector regression, L-1-norm support vector regression, least squares support vector regression, robust least squares support vector regression, and support vector regression with the Huber loss function on both robustness and generalization ability. (C) 2020 Elsevier Inc. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据