%0 Journal Article
%T The Modified BAPG<i><sub>s</i></sub> Method for Support Vector Machine Classifier with Truncated Loss
%A Kexin Ren
%J Applied Mathematics
%P 267-278
%@ 2152-7393
%D 2024
%I Scientific Research Publishing
%R 10.4236/am.2024.154015
%X In this paper, we modify the Bregman APG<i><sub>s</sub></i> (BAPG<i><sub>s</sub></i>) method proposed in (Wang, L, <i>et al.</i>) for solving the support vector machine problem with truncated loss (HTPSVM) given in (Zhu, W, <i>et al.</i>), we also add an adaptive parameter selection technique based on (Ren, K, <i>et al.</i>). In each iteration, we use the linear approximation method to get the explicit solution of the subproblem and set a function <img src=\"https://html.scirp.org//file/7405243-rId12.svg?20240409040807\" /> to apply the Bregman distance. Finally, numerical experiments are performed to verify the efficiency of BAPG<i><sub>s</sub></i>.
%K HTPSVM
%K Bregman Distance
%K BAPG&lt
%K i&gt
%K &lt
%K sub&gt
%K s&lt
%K /i&gt
%K &lt
%K /sub&gt
%K Algorithm
%U http://www.scirp.org/journal/PaperInformation.aspx?PaperID=132373