Semi-Supervised Penalized Output Kernel Regression for Link Prediction
Abstract
Link prediction is addressed as an output kernel learning task through semi-supervised Output Kernel Regression. Working in the framework of RKHS theory with vector-valued functions, we establish a new representer theorem devoted to semi-supervised least square regression. We then apply it to get a new model (POKR: Penalized Output Kernel Regression) and show its relevance using numerical experiments on artificial networks and two real applications using a very low percentage of labeled data in a transductive setting.
Cite
Text
Brouard et al. "Semi-Supervised Penalized Output Kernel Regression for Link Prediction." International Conference on Machine Learning, 2011.Markdown
[Brouard et al. "Semi-Supervised Penalized Output Kernel Regression for Link Prediction." International Conference on Machine Learning, 2011.](https://mlanthology.org/icml/2011/brouard2011icml-semi/)BibTeX
@inproceedings{brouard2011icml-semi,
title = {{Semi-Supervised Penalized Output Kernel Regression for Link Prediction}},
author = {Brouard, Céline and d'Alché-Buc, Florence and Szafranski, Marie},
booktitle = {International Conference on Machine Learning},
year = {2011},
pages = {593-600},
url = {https://mlanthology.org/icml/2011/brouard2011icml-semi/}
}