Analysis of target data-dependent greedy kernel algorithms: Convergence rates for f-, $f \cdot P$- and f/P-greedy

Data-dependent greedy algorithms in kernel spaces are known to provide fast converging interpolants, while being extremely easy to implement and efficient to run. Despite this experimental evidence, no detailed theory has yet been presented. This situation is unsatisfactory especially when compared to the case of the data-independent P -greedy algorithm, for which optimal convergence rates are available, despite its performances being usually inferior to the ones of target data-dependent algorithms. In this work we fill this gap by first defining a new scale of greedy algorithms for interpolation that comprises all the existing ones in a unique analysis, where the degree of dependency of the selection criterion on the functional data is quantified by a real parameter. We then prove new convergence rates where this degree is taken into account and we show that, possibly up to a logarithmic factor, target data-dependent selection strategies provide faster convergence. In particular, for the first time we obtain convergence rates for target data adaptive interpolation that are faster than the ones given by uniform points, without the need of any special assumption on the target function. The rates are confirmed by a number of examples. These results are made possible by a new analysis of greedy algorithms in general Hilbert spaces.

[1]  Robert Schaback,et al.  Error estimates and condition numbers for radial basis function interpolation , 1995, Adv. Comput. Math..

[2]  Bernard Haasdonk,et al.  A novel class of stabilized greedy kernel approximation algorithms: Convergence, stability & uniform point distribution , 2021, J. Approx. Theory.

[3]  Holger Wendland,et al.  Sampling inequalities for sparse grids , 2017, Numerische Mathematik.

[4]  Dirk Pflüger,et al.  Comparison of data-driven uncertainty quantification methods for a carbon dioxide storage benchmark scenario , 2018, Computational Geosciences.

[5]  F. J. Narcowich,et al.  Sobolev Error Estimates and a Bernstein Inequality for Scattered Data Interpolation via Radial Basis Functions , 2006 .

[6]  A. Pinkus.  n-Widths in Approximation Theory , 1985 .

[7]  Ronald A. DeVore,et al.  Some remarks on greedy algorithms , 1996, Adv. Comput. Math..

[8]  G. Santin,et al.  Kernel Methods for Surrogate Modeling , 2019, System- and Data-Driven Methods and Algorithms.

[9]  Gregory E. Fasshauer,et al.  Kernel-based Approximation Methods using MATLAB , 2015, Interdisciplinary Mathematical Sciences.

[10]  M. Urner.  Scattered Data Approximation , 2016 .

[11]  Gregory E. Fasshauer,et al.  Meshfree Approximation Methods with Matlab , 2007, Interdisciplinary Mathematical Sciences.

[12]  R. DeVore,et al.  Orthogonal Matching Pursuit Under the Restricted Isometry Property , 2015, Constructive Approximation.

[13]  Bernard Haasdonk,et al.  Numerical modelling of a peripheral arterial stenosis using dimensionally reduced models and kernel methods , 2018, International journal for numerical methods in biomedical engineering.

[14]  Robert Schaback,et al.  Superconvergence of kernel-based interpolation , 2016, J. Approx. Theory.

[15]  Holger Wendland,et al.  Approximate Interpolation with Applications to Selecting Smoothing Parameters , 2005, Numerische Mathematik.

[16]  Vladimir Temlyakov,et al.  CAMBRIDGE MONOGRAPHS ON APPLIED AND COMPUTATIONAL MATHEMATICS , 2022 .

[17]  Bernard Haasdonk,et al.  Convergence rate of the data-independent P-greedy algorithm in kernel-based approximation , 2016, 1612.02672.

[18]  Ronald DeVore,et al.  Greedy Algorithms for Reduced Bases in Banach Spaces , 2012, Constructive Approximation.

[19]  Bernard Haasdonk,et al.  Greedy Kernel Approximation for Sparse Surrogate Modeling , 2018 .

[20]  Holger Wendland,et al.  Adaptive greedy techniques for approximate solution of large RBF systems , 2000, Numerical Algorithms.

[21]  Robert Schaback,et al.  A Newton basis for Kernel spaces , 2009, J. Approx. Theory.

[22]  S. Mallat,et al.  Adaptive greedy approximations , 1997 .

[23]  Robert Schaback,et al.  Improved error bounds for scattered data interpolation by radial basis functions , 1999, Math. Comput..

[24]  Holger Wendland,et al.  Near-optimal data-independent point locations for radial basis function interpolation , 2005, Adv. Comput. Math..

[25]  Robert Schaback,et al.  Bases for kernel-based spaces , 2011, J. Comput. Appl. Math..

[26]  Emma Perracchione,et al.  A greedy non-intrusive reduced order model for shallow water equations , 2021, J. Comput. Phys..

[27]  R. Schaback,et al.  Numerical Techniques Based on Radial Basis Functions , 2000 .

[28]  Bernard Haasdonk,et al.  Sampling based approximation of linear functionals in Reproducing Kernel Hilbert Spaces , 2021, BIT Numerical Mathematics.

[29]  Holger Wendland,et al.  Sobolev bounds on functions with scattered zeros, with applications to radial basis function surface fitting , 2004, Math. Comput..

[30]  Bernard Haasdonk,et al.  Data-driven surrogates of value functions and applications to feedback control for dynamical systems , 2018 .

[31]  Fred J. Hickernell,et al.  On Dimension-independent Rates of Convergence for Function Approximation with Gaussian Kernels , 2012, SIAM J. Numer. Anal..

[32]  Stéphane Mallat,et al.  Matching pursuits with time-frequency dictionaries , 1993, IEEE Trans. Signal Process..

[33]  Stefan Müller,et al.  Komplexität und Stabilität von kernbasierten Rekonstruktionsmethoden , 2009 .

[34]  Bernard Haasdonk,et al.  A Vectorial Kernel Orthogonal Greedy Algorithm , 2013 .