Learning from Data of Variable Quality
Abstract
We initiate the study of learning from multiple sources of limited data, each of which may be corrupted at a different rate. We develop a com- plete theory of which data sources should be used for two fundamental problems: estimating the bias of a coin, and learning a classifier in the presence of label noise. In both cases, efficient algorithms are provided for computing the optimal subset of data.
Cite
Text
Crammer et al. "Learning from Data of Variable Quality." Neural Information Processing Systems, 2005.Markdown
[Crammer et al. "Learning from Data of Variable Quality." Neural Information Processing Systems, 2005.](https://mlanthology.org/neurips/2005/crammer2005neurips-learning/)BibTeX
@inproceedings{crammer2005neurips-learning,
title = {{Learning from Data of Variable Quality}},
author = {Crammer, Koby and Kearns, Michael and Wortman, Jennifer},
booktitle = {Neural Information Processing Systems},
year = {2005},
pages = {219-226},
url = {https://mlanthology.org/neurips/2005/crammer2005neurips-learning/}
}