Barely Biased Learning for Gaussian Process Regression
Machine Learning
2021-09-21 v1 Machine Learning
Abstract
Recent work in scalable approximate Gaussian process regression has discussed a bias-variance-computation trade-off when estimating the log marginal likelihood. We suggest a method that adaptively selects the amount of computation to use when estimating the log marginal likelihood so that the bias of the objective function is guaranteed to be small. While simple in principle, our current implementation of the method is not competitive computationally with existing approximations.
Cite
@article{arxiv.2109.09417,
title = {Barely Biased Learning for Gaussian Process Regression},
author = {David R. Burt and Artem Artemev and Mark van der Wilk},
journal= {arXiv preprint arXiv:2109.09417},
year = {2021}
}