Abstract:The distribution of centered Gaussian random fields (GRFs) indexed by compacta such as smooth, bounded Euclidean domains or smooth, compact and orientable manifolds is determined by their covariance operators. We consider centered GRFs given as variational solutions to coloring operator equations driven by spatial white noise, with an elliptic self-adjoint pseudodifferential coloring operator from the Hörmander class. This includes the Matérn class of GRFs as a special case. Using biorthogonal multiresolution analyses on the manifold, we prove that the precision and covariance operators, respectively, may be identified with bi-infinite matrices and finite sections may be diagonally preconditioned rendering the condition number independent of the dimension p of this section. We prove that a tapering strategy by thresholding applied on finite sections of the bi-infinite precision and covariance matrices results in optimally numerically sparse approximations. That is, asymptotically only linearly many nonzero matrix entries are sufficient to approximate the original section of the bi-infinite covariance or precision matrix using this tapering strategy to arbitrary precision. The locations of these nonzero matrix entries can be determined a priori. The tapered covariance or precision matrices may also be optimally diagonally preconditioned. Analysis of the relative size of the entries of the tapered covariance matrices motivates novel, multilevel Monte Carlo (MLMC) oracles for covariance estimation, in sample complexity that scales log-linearly with respect to the number p of parameters. In addition, we propose and analyze novel compressive algorithms for simulating and kriging of GRFs. The complexity (work and memory vs. accuracy) of these three algorithms scales near-optimally in terms of the number of parameters p of the sample-wise approximation of the GRF in Sobolev scales.

Uniform approximation of common Gaussian process kernels using equispaced Fourier grids

Uniform approximation of common Gaussian process kernels using equispaced Fourier grids

Efficient Fourier representations of families of Gaussian processes

Compactly-supported nonstationary kernels for computing exact Gaussian processes on big data

Gaussian Process Regression in the Flat Limit

Linear cost and exponentially convergent approximation of Gaussian Matérn processes

On the Approximation of Kernel functions

Multilevel approximation of Gaussian random fields: Covariance compression, estimation, and spatial prediction

Images of Gaussian and other stochastic processes under closed, densely-defined, unbounded linear operators

Contraction rates for conjugate gradient and Lanczos approximate posteriors in Gaussian process regression

Approximation in Hilbert spaces of the Gaussian and related analytic kernels

Gaussian process regression with log-linear scaling for common non-stationary kernels

Fast Adaptive Fourier Integration for Spectral Densities of Gaussian Processes

Vector-valued Gaussian processes on non-Euclidean product spaces: constructive methods and fast simulations based on partial spectral inversion

Integration in reproducing kernel Hilbert spaces of Gaussian kernels

Fast sampling of parameterised Gaussian random fields

Gaussian Process Regression under Computational and Epistemic Misspecification

Further Understanding of a Local Gaussian Process Approximation: Characterising Convergence in the Finite Regime

Exact bounds for some quadratic empirical processes with applications

Gaussian Variational Schemes on Bounded and Unbounded Domains