Evaluating the squared-exponential covariance function in Gaussian processes with integral observations
Machine Learning
2018-12-19 v1 Machine Learning
Abstract
This paper deals with the evaluation of double line integrals of the squared exponential covariance function. We propose a new approach in which the double integral is reduced to a single integral using the error function. This single integral is then computed with efficiently implemented numerical techniques. The performance is compared against existing state of the art methods and the results show superior properties in numerical robustness and accuracy per computation time.
Keywords
Cite
@article{arxiv.1812.07319,
title = {Evaluating the squared-exponential covariance function in Gaussian processes with integral observations},
author = {J. N. Hendriks and C. Jidling and A. Wills and T. B. Schön},
journal= {arXiv preprint arXiv:1812.07319},
year = {2018}
}