Compressed Multiscale Kernel Collocation and Physics-Informed Kernel Learning for Elliptic BVPs

DOI10.48550/ARXIV.2611.02832MaRDI QIDQ7414958arXivFDO

Authors Sara Avesani, Angela Kunoth, Michael Peters, Manel Babay, Michael Multerer

Publication date 2 November 2026

Published in arXiv

Tldr We combine multiscale kernel collocation with samplets and physics-informed learning to efficiently solve elliptic BVPs for hundreds of thousands of unknowns with a novel Schur-complement deflation method for noisy problems.

Full work available at URL https://doi.org/10.48550/ARXIV.2611.02832

arXiv classification

math.NA

Cites work

• A comparison of three explicit local meshless methods using radial basis functions
• A Deflated Version of the Conjugate Gradient Algorithm
• A numerical method for heat transfer problems using collocation and radial basis functions
• Adaptive radial basis function methods for time dependent partial differential equations
• Appending boundary conditions by Lagrange multipliers: Analysis of the LBB condition
• Cambridge Monographs on Applied and Computational Mathematics
• Compactly supported positive definite radial functions
• Compression Techniques for Boundary Integral Equations---Asymptotically Optimal Complexity Estimates
• Convergence of one-level and multilevel unsymmetric collocation for second order elliptic boundary value problems
• Convergence of Unsymmetric Kernel‐Based Meshless Collocation Methods
• Deflation of Conjugate Gradients with Applications to Boundary Value Problems
• Elliptic partial differential equations of second order
• Error estimates and condition numbers for radial basis function interpolation
• Fast wavelet transforms and numerical algorithms I
• Gaussian Processes for Machine Learning
• Interpolation of scattered data: distance matrices and conditionally positive definite functions
• Investigation of regularized techniques for boundary knot method
• Kernel techniques: From machine learning to meshless methods
• Meshless methods: An overview and recent developments
• Monotone Funktionen, Stieltjessche Integrale und harmonische Analyse.
• Multiquadrics - a scattered data approximation scheme with applications to computational fluid-dynamics. I: Surface approximations and partial derivative estimates
• Multiquadrics -- a scattered data approximation scheme with applications to computational fluid-dynamics. II: Solutions to parabolic, hyperbolic and elliptic partial differential equations
• Multiresolution local smoothness detection in non-uniformly sampled multivariate signals
• Multiscale analysis in Sobolev spaces on bounded domains
• Multiscale Bases for the Sparse Representation of Boundary Integral Operators on Complex Geometry
• Multiscale RBF collocation for solving PDEs on spheres
• Multiscale scattered data analysis in samplet coordinates
• Multistep scattered data interpolation using compactly supported radial basis functions
• On unsymmetric collocation by radial basis functions
• Operator-adapted wavelets, fast solvers, and numerical homogenization. From a game theoretic approach to numerical approximation and algorithm design
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Piecewise polynomial, positive definite and compactly supported radial functions of minimal degree
• Pointwise smoothness, two-microlocalization and wavelet coefficients
• RBF multiscale collocation for second-order elliptic boundary value problems
• Samplets: construction and scattered data compression
• Scattered data interpolation on embedded submanifolds with restricted positive definite kernels: Sobolev error estimates
• Solving and Learning Nonlinear PDEs with Gaussian Processes

This page was built for publication: Compressed Multiscale Kernel Collocation and Physics-Informed Kernel Learning for Elliptic BVPs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7414958)