When approximating a black-box function, sampling with active learning focussing on regions with non-linear responses tends to improve accuracy. We present the FLOLA-Voronoi method introduced previously for deterministic responses, and theoretically derive the impact of output uncertainty. The algorithm automatically puts more emphasis on exploration to provide more information to the models.
@article{arxiv.1608.05225,
title = {Active Learning for Approximation of Expensive Functions with Normal Distributed Output Uncertainty},
author = {Joachim van der Herten and Ivo Couckuyt and Dirk Deschrijver and Tom Dhaene},
journal= {arXiv preprint arXiv:1608.05225},
year = {2016}
}