Reproduce “Learning the Entropic Relaxation: Stable Coarse-Grid Emulation with Neural Discrete Equilibrium” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jose Antonio Lara Benitez, Kareem Hegazy, Anastasis Kratsios, Ivan Dokmanić, Maarten V. de Hoop, Michael W. Mahoney
The authors’ NeurDE adds a learned local relaxation parameter to a lattice-Boltzmann solver to restore coarse-grid dissipation while retaining its transport structure. A neural fixed-point update predicts the parameter using entropy-related quantities. At Re 100,000 on a 16×16 grid, NeurDE reproduces expected structures in four training-start seeds, unlike coarse ELBM in one run; on the test start, ELBM performs better.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.