Data Based Construction of Kernels for Semi-Supervised Learning With Less Labels

Data Based Construction of Kernels for Semi-Supervised Learning With Less Labels

This paper deals with the problem of semi-supervised learning using a small number of training samples. Traditional kernel based methods utilize either a fixed kernel or a combination of judiciously chosen kernels from a fixed dictionary. In contrast, we construct a data-dependent kernel utilizing t...

Full description

Bibliographic Details
Main Authors: Hrushikesh Mhaskar, Sergei V. Pereverzyev, Vasyl Yu. Semenov, Evgeniya V. Semenova
Format: Article
Language:English
Published: Frontiers Media S.A. 2019-04-01
Series:Frontiers in Applied Mathematics and Statistics
Subjects:
machine learning
semi-supervised learning
reproducing kernel hilbert spaces
tikhonov regularization
laplace-beltrami operator
gender identification
Online Access:https://www.frontiersin.org/article/10.3389/fams.2019.00021/full

Internet

https://www.frontiersin.org/article/10.3389/fams.2019.00021/full

Similar Items