Margin Based Active Learning
Abstract
We present a framework for margin based active learning of linear separators. We instantiate it for a few important cases, some of which have been previously considered in the literature. We analyze the effectiveness of our framework both in the realizable case and in a specific noisy setting related to the Tsybakov small noise condition.
Cite
Text
Balcan et al. "Margin Based Active Learning." Annual Conference on Computational Learning Theory, 2007. doi:10.1007/978-3-540-72927-3_5Markdown
[Balcan et al. "Margin Based Active Learning." Annual Conference on Computational Learning Theory, 2007.](https://mlanthology.org/colt/2007/balcan2007colt-margin/) doi:10.1007/978-3-540-72927-3_5BibTeX
@inproceedings{balcan2007colt-margin,
title = {{Margin Based Active Learning}},
author = {Balcan, Maria-Florina and Broder, Andrei Z. and Zhang, Tong},
booktitle = {Annual Conference on Computational Learning Theory},
year = {2007},
pages = {35-50},
doi = {10.1007/978-3-540-72927-3_5},
url = {https://mlanthology.org/colt/2007/balcan2007colt-margin/}
}