Capped L2,p-Norm Metric Based on Robust Twin Support Vector Machine with Welsch Loss

A twin bounded support vector machine (TBSVM) is a phenomenon of symmetry that improves the performance traditional classification algorithm. In this paper, we propose an improved model based on TBSVM, called Welsch loss with capped L2,p-norm distance metric robust (WCTBSVM). On one hand, by introducing in problem non-sparse output regularization term solved; thus, generalization and robustness TBSVM principle minimizing structural risk realized. other bounded, smooth, non-convex function introduced to reduce influence noise, which further TBSVM. We use half-quadratic programming algorithm solve non-convexity caused loss. Therefore, WCTBSVM more effective dealing noise compared addition, time complexity speed up convergence algorithm, constructed least squares version WCTBSVM, named fast (FWCTBSVM). Experimental results both UCI artificial datasets show our can better problems.