BibTeX
@ARTICLE{
         Zhou2016ELS,
       author = "Zhou, Xiao Jian and Jiang, Ting",
       title = "Enhancing Least Square Support Vector Regression with Gradient Information",
       journal = "Neural Processing Letters",
       year = "2016",
       volume = "43",
       number = "1",
       pages = "65--83",
       issn = "1573-773X",
       doi = "10.1007/s11063-014-9402-5",
       url = "http://dx.doi.org/10.1007/s11063-014-9402-5",
       abstract = "Traditional methods of constructing of least square support vector regression
         (LSSVR) do not consider the gradients of the true function but just think about the exact responses
         at samples. If gradient information is easy to get, it should be used to enhance the surrogate. In
         this paper, the gradient-enhanced least square support vector regression (GELSSVR) is developed with
         a direct formulation by incorporating gradient information into the traditional LSSVR. The
         efficiencies of this technique are compared by analytical function fitting and two real life
         problems (the recent U.S. actuarial life table and Borehole). The results show that GELSSVR provides
         more reliable prediction results than LSSVR alone.",
       ad_area = "Machine Learning",
       ad_tools = "ADiMat"
}