A Closer Look at Covering Number Bounds for Gaussian Kernels

Abstract

We establish some new bounds on the log-covering numbers of (anisotropic) Gaussian reproducing kernel Hilbert spaces. Unlike previous results in this direction we focus on small explicit constants and their dependency on crucial parameters such as the kernel bandwidth and the size and dimension of the underlying space.

accepted manuscript in J. Complexity

References in corpus (2)

• Adaptive Bayesian estimation using a Gaussian random field with inverse Gamma bandwidth
• Gaussian Processes and Kernel Methods: A Review on Connections and Equivalences

Cited by in corpus (4)

• A close look at the entropy numbers of the unit ball of the Reproducing Hilbert Space of isotropic positive definite kernels
• Generalized Kernel Thinning
• Interpolation and Learning with Scale Dependent Kernels
• Approximation in Hilbert spaces of the Gaussian and related analytic kernels