Reproduce “Adjoint-aided inference of Gaussian process driven differential equations” 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
Paterne Gahungu, Christopher W Lanyon, Mauricio A Alvarez, Engineer Bainomugisha, Michael Smith, Richard D. Wilkinson
The authors infer an unknown forcing from noisy observations of a linear system using adjoints. They represent the forcing with finite basis functions and use observation-specific adjoints to construct a linear model, yielding a Gaussian posterior under Gaussian priors and errors. In a simulated PDE case, results match MCMC at printed precision using 75 adjoint solves; exactness applies only to the chosen basis.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.