Reproduce “Bayesian Physics-Informed Extreme Learning Machine for Forward and Inverse PDE Problems with Noisy Data” 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
Xu Liu, Wen Yao, Wei Peng, Weien Zhou
BPIELM combines a fixed-random-feature physics-informed solver with Bayesian inference over output coefficients, producing distributions over solutions and inverse PDE parameters. In synthetic noisy tests on linear PDEs, it reports uncertainty and is less sensitive to neuron count than PIELM. For a noisy Poisson example, its MAE was 0.019 versus 0.029 for PIELM; findings are limited to tested linear problems.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.