Selecting Informative Universum Sample for Semi-Supervised Learning
Proceedings of the Twenty-First International Joint Conference on Artificial Intelligence. Pages 1016-1020.
Abstract
The Universum sample, which is defined as the sample that doesn't belong to any of the classes the learning task concerns, has been proved to be helpful in both supervised and semi-supervised settings. The former works treat the Universum samples equally. Our research found that not all the Universum samples are helpful, and we propose a method to pick the informative ones, i.e., in-between Universum samples. We also set up a new semi-supervised framework to incorporate the in-between Universum samples. Empirical experiments show that our method outperforms the former ones.Shuo Chen, Changshui Zhang
@inproceedings{ijcai2009p172,
title = {Selecting Informative Universum Sample for Semi-Supervised Learning},
author = {Chen, Shuo and Zhang, Changshui},
booktitle = {Proceedings of the Twenty-First International Joint Conference on
Artificial Intelligence, {IJCAI-09}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Hiroaki Kitano},
pages = {1016--1020},
year = {2009},
url = {https://ijcai.org/proceedings/2009/172},
}
TY - CPAPER TI - Selecting Informative Universum Sample for Semi-Supervised Learning AU - Chen, Shuo AU - Zhang, Changshui T2 - Proceedings of the Twenty-First International Joint Conference on Artificial Intelligence, IJCAI-09 PB - International Joint Conferences on Artificial Intelligence PY - 2009 SP - 1016 EP - 1020 UR - https://ijcai.org/proceedings/2009/172 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
