On the Correspondence between Gaussian Processes and Geometric Harmonics
Machine Learning
2021-10-07 v1 Machine Learning
Optimization and Control
Spectral Theory
Abstract
We discuss the correspondence between Gaussian process regression and Geometric Harmonics, two similar kernel-based methods that are typically used in different contexts. Research communities surrounding the two concepts often pursue different goals. Results from both camps can be successfully combined, providing alternative interpretations of uncertainty in terms of error estimation, or leading towards accelerated Bayesian Optimization due to dimensionality reduction.
Cite
@article{arxiv.2110.02296,
title = {On the Correspondence between Gaussian Processes and Geometric Harmonics},
author = {Felix Dietrich and Juan M. Bello-Rivas and Ioannis G. Kevrekidis},
journal= {arXiv preprint arXiv:2110.02296},
year = {2021}
}
Comments
26 pages, 9 figures