Active Learning for Probability Estimation Using Jensen-Shannon Divergence
Abstract
Active selection of good training examples is an important approach to reducing data-collection costs in machine learning; however, most existing methods focus on maximizing classification accuracy. In many applications, such as those with unequal misclassification costs, producing good class probability estimates (CPEs) is more important than optimizing classification accuracy. We introduce novel approaches to active learning based on the algorithms Bootstrap-LV and ActiveDecorate , by using Jensen-Shannon divergence (a similarity measure for probability distributions) to improve sample selection for optimizing CPEs. Comprehensive experimental results demonstrate the benefits of our approaches.
Cite
Text
Melville et al. "Active Learning for Probability Estimation Using Jensen-Shannon Divergence." European Conference on Machine Learning, 2005. doi:10.1007/11564096_28Markdown
[Melville et al. "Active Learning for Probability Estimation Using Jensen-Shannon Divergence." European Conference on Machine Learning, 2005.](https://mlanthology.org/ecmlpkdd/2005/melville2005ecml-active/) doi:10.1007/11564096_28BibTeX
@inproceedings{melville2005ecml-active,
title = {{Active Learning for Probability Estimation Using Jensen-Shannon Divergence}},
author = {Melville, Prem and Yang, Stewart M. and Saar-Tsechansky, Maytal and Mooney, Raymond J.},
booktitle = {European Conference on Machine Learning},
year = {2005},
pages = {268-279},
doi = {10.1007/11564096_28},
url = {https://mlanthology.org/ecmlpkdd/2005/melville2005ecml-active/}
}