Reproduce “A Probabilistic Framework for Solving High-Frequency Helmholtz Equations via Diffusion Models” 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
Yicheng Zou, Samuel Lanthaler, Hossein Salahshoor
The authors develop a conditional diffusion model to predict distributions of two-dimensional steady-state Helmholtz wavefields from sound-speed maps and source information. Across six synthetic frequencies, it reports lower relative L-squared, H-one, and energy errors than three deterministic baselines; at 2.5 MHz, L-squared error is 0.095 ± 0.019 versus FNO’s 0.412. The comparison is limited to this benchmark and a 1,000-step sampler.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.