Reproduce “Solving and Learning Navier-Stokes Dynamics with the Neural Basis Method” 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
Yuhe Wang, Zhu Wang, Min Wang
The Neural Basis Operator represents incompressible flow with fixed divergence-free neural bases and predicts their coefficients from physical parameters and time. It removes pressure-gradient contributions through projection and trains by minimizing the projected equation residual. For Kolmogorov flow, NBO-RC reports 0.45% relative L² velocity error, versus 1.3% for NBO-DD; this comparison is benchmark-specific, not universal.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.