Residual-Christoffel Sampling for Random Feature Collocation of Linear PDEs

DOI10.48550/arXiv.2607.13382MaRDI QIDQ7284056arXivFDO

Authors Jiale Linghu, Yangshuai Wang

Publication date 15 July 2026

Published in arXiv

Tldr Random feature collocation method that constructs an operator-aware discretization to stabilize the linear least-squares system by sampling the residual equations with respect to Christoffel density. This results in whitened interior Gram matrix that approximates the reference Gram on the retained residual space, allowing for reduced condition numbers and iteration counts in numerical experiments.

Full work available at URL https://doi.org/10.48550/arXiv.2607.13382

arXiv classification

math.NAphysics.comp-ph

Cites work

• A Christoffel function weighted least squares algorithm for collocation approximations
• A comprehensive study of non-adaptive and residual-based adaptive sampling for physics-informed neural networks
• A uniformly accurate multiscale time integrator pseudospectral method for the Dirac equation in the nonrelativistic limit regime
• Adaptive meshfree approximation for linear elliptic partial differential equations with PDE-greedy kernel methods
• An extreme learning machine-based method for computational PDEs in higher dimensions
• An introduction to matrix concentration inequalities
• Approximation of eigenfunctions in kernel-based spaces
• Blendenpik: Supercharging LAPACK's Least-Squares Solver
• Bridging traditional and machine learning-based algorithms for solving PDEs: the random feature method
• Christoffel Adaptive Sampling for Sparse Random Feature Expansions
• Coherence motivated sampling and convergence analysis of least squares polynomial chaos regression
• Computing the ground state and dynamics of the nonlinear Schrödinger equation with nonlocal interactions via the nonuniform FFT
• Concentration inequalities and moment bounds for sample covariance operators
• CS4ML: A general framework for active learning with arbitrary data based on Christoffel functions
• DeepXDE: a deep learning library for solving differential equations
• Eigenvalues of Analytic Kernels
• Failure-Informed Adaptive Sampling for PINNs
• High-precision randomized preconditioned iterative methods for the random feature method
• Input sparsity time low-rank approximation via ridge leverage score sampling
• Local Extreme Learning Machines and Domain Decomposition for Solving Linear and Nonlinear Partial Differential Equations
• Local Feature Filtering for Scalable and Well-Conditioned Domain-Decomposed Random Feature Methods
• Local Randomized Neural Networks with Discontinuous Galerkin Methods for Partial Differential Equations
• LSRN: A parallel iterative solver for strongly over- or underdetermined systems
• Multiquadrics -- a scattered data approximation scheme with applications to computational fluid-dynamics. II: Solutions to parabolic, hyperbolic and elliptic partial differential equations
• Numerical methods and comparison for the Dirac equation in the nonrelativistic limit regime
• On unsymmetric collocation by radial basis functions
• Optimal sampling for least-squares approximation
• Optimal weighted least-squares methods
• The Random Feature Method for Time-Dependent Problems
• User-friendly tail bounds for sums of random matrices

This page was built for publication: Residual-Christoffel Sampling for Random Feature Collocation of Linear PDEs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7284056)