Active Learning for Optimal Minimization of Experimental Characterization Uncertainty
Abstract
Collecting experimental measurements is rarely an end in itself; rather, measurements inform key outcome statistics. Standard active learning procedures can drive a cumulative decrease in measurement uncertainty, but do not account for the uncertainty of the outcome. Here we present an active learning framework that operates to minimize the uncertainty of the outcome, and demonstrate its applicability with imaging and spectroscopic tasks. We show how our framework can effectively select regions for measurement without iteratively retraining a model. We conclude with two instances where our framework has outperformed standard active learning procedures to accelerate the classification of unknown samples.
Cite
Text
Schwarting et al. "Active Learning for Optimal Minimization of Experimental Characterization Uncertainty." NeurIPS 2024 Workshops: BDU, 2024.Markdown
[Schwarting et al. "Active Learning for Optimal Minimization of Experimental Characterization Uncertainty." NeurIPS 2024 Workshops: BDU, 2024.](https://mlanthology.org/neuripsw/2024/schwarting2024neuripsw-active/)BibTeX
@inproceedings{schwarting2024neuripsw-active,
title = {{Active Learning for Optimal Minimization of Experimental Characterization Uncertainty}},
author = {Schwarting, Marcus and Seifert, Nathan and Ward, Logan and Blaiszik, Ben and Foster, Ian and Chen, Yuxin and Prozument, Kirill},
booktitle = {NeurIPS 2024 Workshops: BDU},
year = {2024},
url = {https://mlanthology.org/neuripsw/2024/schwarting2024neuripsw-active/}
}