文献类型：期刊文献

英文题名：Convergence rate of semi-supervised gradient learning algorithms

作者：Sheng, Baohuai[1];Xiang, Daohong[2];Ye, Peixin[3,4]

机构：[1]Shaoxing Univ, Dept Math, Shaoxing 312000, Zhejiang, Peoples R China;[2]Zhejiang Normal Univ, Dept Math, Jinhua 312004, Peoples R China;[3]Nankai Univ, Sch Math, Tianjin 312000, Peoples R China;[4]Nankai Univ, LPMC, Tianjin 312000, Peoples R China

年份：2015

卷号：13

期号：4

外文期刊名：INTERNATIONAL JOURNAL OF WAVELETS MULTIRESOLUTION AND INFORMATION PROCESSING

收录：SCI-EXPANDED(收录号：WOS:000358621600003)、、EI(收录号：20152400941166)、Scopus(收录号：2-s2.0-84938414548)、WOS

基金：This work is supported by the National Natural Science Foundation of China under Grants Nos. 61179041, 11471292 and 11271199. The authors thank the reviewers for giving many valuable suggestions and comments which make the paper presented in a better form.

语种：英文

外文关键词：Semi-supervised learning; convergence rate; reproducing kernel Hilbert space

外文摘要：Semi-supervised learning deals with learning with a small amount labeled sample and a large amount of unlabeled sample to improve the learning ability. The purpose of the semi-supervised gradient learning is to increase the smoothness of the solution using unlabeled gradient data. In this paper, we study the semi-supervised kernel-based regularization scheme involving function gradient value. We show that the learning rate can be bounded by a K-functional with gradients of the function, which verify how the unlabeled gradient data quantitatively influences the learning rate. Some approaches from convex analysis play a key role in our error analysis.