Get Started
Home
Topics
Search
Library
Research questionHow can mesh-free differential operators stay accurate at high resolvable wavenumbers on disordered nodes?Mesh-free stencils can contain more unknown weights than polynomial-consistency constraints, leaving some degrees of freedom unspecified. Consistency mainly controls low-wavenumber behavior and does not ensure accurate responses for fine-scale resolvable content.
Machine Learning
Neural and Evolutionary Computing
Research Paper
Technology
Latest papersRecent research connected to this question, newest first.Learning Spectral-Like Mesh-Free DiscretisationsThe source considers SPH with kernel corrections, RBF-FD, and LABFM, and presents a neural parameterization of stencil weights conditioned on local node geometry with an exact consistency projection. Evidence is limited to modal analysis on disordered two-dimensional node distributions, including a learned fourth-order operator, comparisons with LABFM and structured-grid finite differences, and refinement-rate measurements.research paper · Sep 2, 2026
Related questions
How can DeepONet-style neural operators represent sharp moving discontinuities in shock-dominated, low-viscosity PDEs?How can smooth density estimation preserve accuracy under relaxed local differential privacy?How can random-feature spaces achieve spectral accuracy for multidimensional targets and elliptic PDEs while controlling ill-conditioning?How can neural signed distance fields yield exact zero-level-set meshes without losing high-frequency geometry?