Multiscale Approximation and Reproducing Kernel Hilbert Space Methods

  title={Multiscale Approximation and Reproducing Kernel Hilbert Space Methods},
  author={Michael Griebel and Christian Rieger and Barbara Zwicknagl},
  journal={SIAM J. Numer. Anal.},
We consider reproducing kernels $K:\Omega\times \Omega \to \mathbb{R}$ in multiscale series expansion form, i.e., kernels of the form $K\left(\boldsymbol{x},\boldsymbol{y}\right)=\sum_{\ell\in\mathbb{N}}\lambda_\ell\sum_{j\in I_\ell}\phi_{\ell,j}\left(\boldsymbol{x}\right)\phi_{\ell,j}\left(\boldsymbol{y}\right)$ with weights $\lambda_\ell$ and structurally simple basis functions $\left\{\phi_{\ell,i}\right\}$. Here, we deal with basis functions such as polynomials or frame systems, where, for… 

An introduction to the Hilbert-Schmidt SVD using iterated Brownian bridge kernels

A class of so-called iterated Brownian bridge kernels which allow the discussion to keep the discussion as simple and accessible as possible are introduced.

An inverse theorem for compact Lipschitz regions in ℝd using localized kernel bases

Inverse estimates for finite dimensional spaces arising in radial basis function approximation and meshless methods are developed that consider control Sobolev norms of linear combinations of a localized basis by the $L_p$ norm over a bounded domain.

A Forward Backward Greedy approach for Sparse Multiscale Learning

  • P. ShekharA. Patra
  • Computer Science
    Computer Methods in Applied Mechanics and Engineering
  • 2022

Direct and Inverse Results on Bounded Domains for Meshless Methods via Localized Bases on Manifolds

This article develops direct and inverse estimates for certain finite dimensional spaces arising in kernel approximation. Both the direct and inverse estimates are based on approximation spaces

Regularized Kernel-Based Reconstruction in Generalized Besov Spaces

A theoretical framework for reproducing kernel-based reconstruction methods in certain generalized Besov spaces based on positive, essentially self-adjoint operators is presented and explicit coupling relations between the series truncation, the regularization parameters and the data set are derived.

An inverse theorem on bounded domains for meshless methods using localized bases

This article develops inverse estimates for finite dimensional spaces arising in kernel approximation and meshless methods. These control Sobolev norms of linear combinations of a localized basis by

An improved radial basis functions method for the high-order Volterra–Fredholm integro-differential equations

This paper is aimed at rectifying the numerical solution of linear Volterra–Fredholm integro-differential equations with the method of radial basis functions (RBFs). In this method, the spectral

Existence of Unique Solutions to the Telegraph Equation in Binary Reproducing Kernel Hilbert Spaces

We demonstrate the existence of a unique solution to a nonhomogeneous telegraph initial/boundary value problem on the unit square in an appropriate binary reproducing kernel Hilbert space which

On the Convergence Rate of Sparse Grid Least Squares Regression

This paper presents a framework which will allow for a thorough theoretical analysis of stability properties, error decay behavior and appropriate couplings between the dataset size and the grid size and rigorously derive upper bounds on the expected error for sparse grid least squares regression.



Extension of sampling inequalities to Sobolev semi-norms of fractional order and derivative data

The extensions established in this paper allow us to bound fractional order semi-norms and to incorporate, if available, values of partial derivatives on the discrete set.

An extension of a bound for functions in Sobolev spaces, with applications to (m, s)-spline interpolation and smoothing

The Sobolev bound is applied to derive error estimates for interpolating and smoothing (m, s)-splines in the case of smoothing and noisy data as well as exact data are considered.

Tight frame expansions of multiscale reproducing kernels in Sobolev spaces

Stable Evaluation of Gaussian Radial Basis Function Interpolants

We provide a new way to compute and evaluate Gaussian radial basis function interpolants in a stable way with a special focus on small values of the shape parameter, i.e., for “flat” kernels. This

Multiscale kernels

The reconstruction uses standard methods of interpolation by positive definite reproducing kernels in Hilbert spaces, but it adopts techniques from wavelet theory and shift–invariant spaces to construct a new class of kernels as multiscale superpositions of shifts and scales of a single compactly supported function φ.

Sobolev Error Estimates and a Bernstein Inequality for Scattered Data Interpolation via Radial Basis Functions

Error estimates for scattered-data interpolation via radial basis functions (RBFs) for target functions in the associated reproducing kernel Hilbert space (RKHS) have been known for a long time.

Frames and Space Splittings in Hilbert Spaces

This is the rst part of lecture notes based on short courses held by the author at the Universities of Bonn and Lancaster in We review the abstract theory of frames and their generalization the so

On Learning with Integral Operators

A technique based on a concentration inequality for Hilbert spaces is used to provide new much simplified proofs for a number of results in spectral approximation of graph Laplacian operator extending and strengthening results from von Luxburg et al. (2008).

Sampling and Stability

This survey first describes these two types of inequalities in general and shows how to derive a stability inequality from a sampling inequality plus an inverse inequality on a finite–dimensional trial space.