Asymptotic Bounds for Smoothness Parameter Estimates in Gaussian Process Interpolation
share
It is common to model a deterministic response function, such as the output of a computer experiment, as a Gaussian process with a Matérn covariance kernel. The smoothness parameter of a Matérn kernel determines many important properties of the model in the large data limit, such as the rate of convergence of the conditional mean to the response function. We prove that the maximum likelihood and cross-validation estimates of the smoothness parameter cannot asymptotically undersmooth the truth when the data are obtained on a fixed bounded subset of ℝ^d. That is, if the data-generating response function has Sobolev smoothness ν_0 + d/2, then the smoothness parameter estimates cannot remain below ν_0 as more data are obtained. The results are based on approximation theory in Sobolev spaces and a general theorem, proved using reproducing kernel Hilbert space techniques, about sets of values the parameter estimates cannot take.
READ FULL TEXT