The Modified BAPG<i><sub>s</i></sub> Method for Support Vector Machine Classifier with Truncated Loss

Abstract

In this paper, we modify the Bregman APG s (BAPG s ) method proposed in (Wang, L, et al.) for solving the support vector machine problem with truncated loss (HTPSVM) given in (Zhu, W, et al.), we also add an adaptive parameter selection technique based on (Ren, K, et al.). In each iteration, we use the linear approximation method to get the explicit solution of the subproblem and set a function φ to apply the Bregman distance. Finally, numerical experiments are performed to verify the efficiency of BAPG s .