Abstract:Gaussian process (GP) regression is a fundamental tool in Bayesian statistics. It is also known as kriging and is the Bayesian counterpart to the frequentist kernel ridge regression. Most of the theoretical work on GP regression has focused on a large-$n$ asymptotics, i.e. as the amount of data increases. Fixed-sample analysis is much more difficult outside of simple cases, such as locations on a regular grid. In this work we perform a fixed-sample analysis that was first studied in the context of approximation theory by Driscoll & Fornberg (2002), called the ``flat limit''. In flat-limit asymptotics, the goal is to characterise kernel methods as the length-scale of the kernel function tends to infinity, so that kernels appear flat over the range of the data. Surprisingly, this limit is well-defined, and displays interesting behaviour: Driscoll & Fornberg showed that radial basis interpolation converges in the flat limit to polynomial interpolation, if the kernel is Gaussian. Subsequent work showed that this holds true in the multivariate setting as well, but that kernels other than the Gaussian may have (polyharmonic) splines as the limit interpolant. Leveraging recent results on the spectral behaviour of kernel matrices in the flat limit, we study the flat limit of Gaussian process regression. Results show that Gaussian process regression tends in the flat limit to (multivariate) polynomial regression, or (polyharmonic) spline regression, depending on the kernel. Importantly, this holds for both the predictive mean and the predictive variance, so that the posterior predictive distributions become equivalent. Our results have practical consequences: for instance, they show that optimal GP predictions in the sense of leave-one-out loss may occur at very large length-scales, which would be invisible to current implementations because of numerical difficulties.

Easy representation of multivariate functions with low-dimensional terms via Gaussian process regression kernel design: applications to machine learning of potential energy surfaces and kinetic energy densities from sparse data

The loss of the property of locality of the kernel in high-dimensional Gaussian process regression on the example of the fitting of molecular potential energy surfaces

Gaussian Process Regression under Computational and Epistemic Misspecification

Efficient Two-Stage Gaussian Process Regression Via Automatic Kernel Search and Subsampling

High-Dimensional Gaussian Process Regression with Soft Kernel Interpolation

Manifold Gaussian Processes for Regression

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

Thin and Deep Gaussian Processes

Sparsity-Aware Distributed Learning for Gaussian Processes with Linear Multiple Kernel

Degeneration of kernel regression with Matern kernels into low-order polynomial regression in high dimension

Mixture of Gaussian Processes and its Applications

Polynomial Representation of the Gaussian Process

A Unifying Perspective on Non-Stationary Kernels for Deeper Gaussian Processes

Gaussian Process Regression in the Flat Limit

Gaussian Processes with Spectral Delta kernel for higher accurate Potential Energy surfaces for large molecules

Longitudinal Deep Kernel Gaussian Process Regression

Solving High Frequency and Multi-Scale PDEs with Gaussian Processes

Scaling Gaussian Process Regression with Derivatives

A hybrid data driven-physics constrained Gaussian process regression framework with deep kernel for uncertainty quantification

Vecchia Gaussian Processes: Probabilistic Properties, Minimax Rates and Methodological Developments