Spectral representations of interpolation spaces of reproducing kernel Hilbert spaces

2025/08/22 by Bitzer, Michael, Steinwart, Ingo
#46E22 (Primary) 46B70 #47A70 #47B32 #47B34 #47G10 #68T05 (Secondary) #FOS: Mathematics #Functional Analysis (math.FA)

paper · doi:10.48550/arxiv.2508.16492

Abstract

In statistical learning theory, interpolation spaces of the form [L²,H]_θ,r, where H is a reproducing kernel Hilbert space, are in widespread use. So far, however, they are only well understood for fine index r=2. We generalise existing results from r=2 to all possible values of r. In particular, we present a spectral decomposition of such spaces, analyse their embedding properties, and describe connections to the theory of Banach spaces of functions. We additionally present example applications of our results to regularisation error estimation in statistical learning.

Citations

• Beyond ReLU: How Activations Affect Neural Kernels and Random Wide Networks
• Gaussian Processes and Reproducing Kernels: Connections and Equivalences
• General superconvergence for kernel-based approximation
• When does a Gaussian process have its paths in a reproducing kernel Hilbert space?
• Towards Optimal Sobolev Norm Rates for the Vector-Valued Regularized Least-Squares Algorithm
• On the Asymptotic Learning Curves of Kernel Ridge Regression under Power-law Decay
• Physics-Informed Gaussian Process Regression Generalizes Linear PDE Solvers
• Sobolev Spaces, Kernels and Discrepancies over Hyperspheres
• Optimal Rates for Regularized Conditional Mean Embedding Learning
• Disintegration of Gaussian Measures for Sequential Assimilation of Linear Operator Data
• Understanding neural networks with reproducing kernel Banach spaces
• Small Sample Spaces for Gaussian Processes
• Neural Tangent Kernel: Convergence and Generalization in Neural Networks
• Convergence Types and Rates in Generic Karhunen-Loève Expansions with Applications to Sample Path Properties

Related

• Sparse Representer Theorems for Learning in Reproducing Kernel Banach Spaces
• Consistent Interpolating Ensembles via the Manifold-Hilbert Kernel
• Reproducing Kernel Hilbert Spaces for Penalized Regression: A Tutorial
• Reproducing Kernel Banach Spaces with the l1 Norm II: Error Analysis for Regularized Least Square Regression
• When is there a Representer Theorem? Nondifferentiable Regularisers and Banach spaces