An effective way to integrate ε-support vector regression with gradients

作者：

Highlights：

• The gradient information is incorporated into the construction of the traditional SVR model.

• The gradient-enhanced SVR is developed by adding the gradient information into the kernel matrix.

• The gradient-enhanced SVR outperforms the traditional SVR model.

摘要

•The gradient information is incorporated into the construction of the traditional SVR model.•The gradient-enhanced SVR is developed by adding the gradient information into the kernel matrix.•The gradient-enhanced SVR outperforms the traditional SVR model.

论文关键词：ε-support vector regression,Metamodel,Gradient information,Machine learning

论文评审过程：Received 6 August 2017, Revised 16 January 2018, Accepted 22 January 2018, Available online 3 February 2018, Version of Record 3 February 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.01.036