Reproduce “Noise, Denoise, Correct: MCMC Posterior Sampling with Diffusion Priors in Three Steps” 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
So Takao, Gregory David Bellchambers, Luke Ye, Sanmitra Ghosh, Michalis Michaelides
Diffusion Waltz estimates initial vorticity from noisy, downsampled observations after Navier–Stokes evolution. It perturbs a candidate, denoises it into a plausible field, and checks acceptance; EKCN first nudges candidates toward observations using ensemble information. In one high-noise, high-nonlinearity test, DW-EKCN achieved mean CRPS 0.207 versus Blade’s 0.231 across five unseen examples; its interacting-ensemble implementation is not exact.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.