Active Surrogate Estimators: An Active Learning Approach to Label-Efficient Model Evaluation
Machine Learning
2022-10-20 v2 Machine Learning
Abstract
We propose Active Surrogate Estimators (ASEs), a new method for label-efficient model evaluation. Evaluating model performance is a challenging and important problem when labels are expensive. ASEs address this active testing problem using a surrogate-based estimation approach that interpolates the errors of points with unknown labels, rather than forming a Monte Carlo estimator. ASEs actively learn the underlying surrogate, and we propose a novel acquisition strategy, XWED, that tailors this learning to the final estimation task. We find that ASEs offer greater label-efficiency than the current state-of-the-art when applied to challenging model evaluation problems for deep neural networks.
Keywords
Cite
@article{arxiv.2202.06881,
title = {Active Surrogate Estimators: An Active Learning Approach to Label-Efficient Model Evaluation},
author = {Jannik Kossen and Sebastian Farquhar and Yarin Gal and Tom Rainforth},
journal= {arXiv preprint arXiv:2202.06881},
year = {2022}
}
Comments
Accepted for publication at NeurIPS 2022