Pointwise Uncertainty Quantification for Sparse Variational Gaussian Process Regression with a Brownian Motion Prior
Abstract
We study pointwise estimation and uncertainty quantification for a sparse variational Gaussian process method with eigenvector inducing variables. For a rescaled Brownian motion prior, we derive theoretical guarantees and limitations for the frequentist size and coverage of pointwise credible sets. For sufficiently many inducing variables, we precisely characterize the asymptotic frequentist coverage, deducing when credible sets from this variational method are conservative and when overconfident/misleading. We numerically illustrate the applicability of our results and discuss connections with other common Gaussian process priors.
Cite
Text
Travis and Ray. "Pointwise Uncertainty Quantification for Sparse Variational Gaussian Process Regression with a Brownian Motion Prior." Neural Information Processing Systems, 2023.Markdown
[Travis and Ray. "Pointwise Uncertainty Quantification for Sparse Variational Gaussian Process Regression with a Brownian Motion Prior." Neural Information Processing Systems, 2023.](https://mlanthology.org/neurips/2023/travis2023neurips-pointwise/)BibTeX
@inproceedings{travis2023neurips-pointwise,
title = {{Pointwise Uncertainty Quantification for Sparse Variational Gaussian Process Regression with a Brownian Motion Prior}},
author = {Travis, Luke and Ray, Kolyan},
booktitle = {Neural Information Processing Systems},
year = {2023},
url = {https://mlanthology.org/neurips/2023/travis2023neurips-pointwise/}
}