Finite Element Representations of Gaussian Processes: Balancing Numerical and Statistical Accuracy

09/06/2021
by   Daniel Sanz-Alonso, et al.
0

The stochastic partial differential equation approach to Gaussian processes (GPs) represents Matérn GP priors in terms of n finite element basis functions and Gaussian coefficients with sparse precision matrix. Such representations enhance the scalability of GP regression and classification to datasets of large size N by setting n≈ N and exploiting sparsity. In this paper we reconsider the standard choice n ≈ N through an analysis of the estimation performance. Our theory implies that, under certain smoothness assumptions, one can reduce the computation and memory cost without hindering the estimation accuracy by setting n ≪ N in the large N asymptotics. Numerical experiments illustrate the applicability of our theory and the effect of the prior lengthscale in the pre-asymptotic regime.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/19/2022

Error analysis for a statistical finite element method

The recently proposed statistical finite element (statFEM) approach synt...
research
03/10/2018

How to solve the stochastic partial differential equation that gives a Matérn random field using the finite element method

This tutorial teaches parts of the finite element method (FEM), and solv...
research
12/07/2020

Deep Gaussian Processes for geophysical parameter retrieval

This paper introduces deep Gaussian processes (DGPs) for geophysical par...
research
02/24/2014

Manifold Gaussian Processes for Regression

Off-the-shelf Gaussian Process (GP) covariance functions encode smoothne...
research
12/10/2021

Numerical methods for Mean field Games based on Gaussian Processes and Fourier Features

In this article, we propose two numerical methods, the Gaussian Process ...
research
03/09/2023

Reconstructing the Hubble parameter with future Gravitational Wave missions using Machine Learning

We study the prospects of Machine Learning algorithms like Gaussian proc...
research
06/11/2021

Measuring the sensitivity of Gaussian processes to kernel choice

Gaussian processes (GPs) are used to make medical and scientific decisio...

Please sign up or login with your details

Forgot password? Click here to reset