Uniform approximation of common Gaussian process kernels using equispaced Fourier grids
From MaRDI portal
Abstract: The high efficiency of a recently proposed method for computing with Gaussian processes relies on expanding a (translationally invariant) covariance kernel into complex exponentials, with frequencies lying on a Cartesian equispaced grid. Here we provide rigorous error bounds for this approximation for two popular kernels -- Mat'ern and squared exponential -- in terms of the grid spacing and size. The kernel error bounds are uniform over a hypercube centered at the origin. Our tools include a split into aliasing and truncation errors, and bounds on sums of Gaussians or modified Bessel functions over various lattices. For the Mat'ern case, motivated by numerical study, we conjecture a stronger Frobenius-norm bound on the covariance matrix error for randomly-distributed data points. Lastly, we prove bounds on, and study numerically, the ill-conditioning of the linear systems arising in such regression problems.
Recommendations
Cited in
(5)- Error bounds and the asymptotic setting in kernel-based approximation
- A gradient-based and determinant-free framework for fully Bayesian Gaussian process regression
- Gaussian process regression with log-linear scaling for common non-stationary kernels
- Efficient Fourier representations of families of Gaussian processes
- Equispaced Fourier representations for efficient Gaussian process regression from a billion data points
This page was built for publication: Uniform approximation of common Gaussian process kernels using equispaced Fourier grids
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6437112)
