首页|Investigators from Shenzhen University Zero in on Support Vector Machines (Robust Twin Bounded Support Vector Classifier With Manifold Regularization)
Investigators from Shenzhen University Zero in on Support Vector Machines (Robust Twin Bounded Support Vector Classifier With Manifold Regularization)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News - Investigators publish new report on Ma chine Learning - Support Vector Machines.According to news reporting originatin g from Shenzhen, People’s Republic of China, by NewsRx correspondents,research stated, “Support vector machine (SVM), as a supervised learning method, has different kinds of varieties with significant performance. In recent years, more res earch focused on nonparallelSVM, where twin SVM (TWSVM) is the typical one.” Funders for this research include National Natural Science Foundation of China ( NSFC), ShenzhenMunicipal Science and Technology Innovation Council.Our news editors obtained a quote from the research from Shenzhen University, “I n order to reducethe influence of outliers, more robust distance measurements a re considered in these methods, but thediscriminability of the models is neglec ted. In this article, we propose robust manifold twin bounded SVM(RMTBSVM), whi ch considers both robustness and discriminability. Specifically, a novel norm, t hat is,capped L-1-norm, is used as the distance metric for robustness, and a ro bust manifold regularization isadded to further improve the robustness and clas sification performance. In addition, we also use the kernelmethod to extend the proposed RMTBSVM for nonlinear classification. We introduce the optimization problems of the proposed model. Subsequently, effective algorithms for both linear and nonlinear cases areproposed and proved to be convergent. Moreover, the exp eriments are conducted to verify the effectivenessof our model.”
ShenzhenPeople’s Republic of ChinaAsiaMachine LearningSupport Vector MachinesShenzhen University
NETL
NSTL
