Reproduce “Stable and Calibrated Latent Force ODE Filters” 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
Paul Fischer, Jonathan Schmidt, Philipp Hennig
The authors propose selecting state diffusion in a latent-force ODE filter according to how much observations inform the hidden force, conditional on the differential equation. Across 60 simulated problems in three nonlinear systems, this criterion misses none under the paper’s force-error threshold, while observation evidence misses four. The classical latent-force model remains competitive; the result is not a guarantee.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.