Reproduce “Robust Filtering and Smoothing with Gaussian Processes” 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
Marc Peter Deisenroth, Ryan Turner, Marco F. Huber, Uwe D. Hanebeck, Carl Edward Rasmussen
GP-RTSS smooths hidden states with learned Gaussian process transition and measurement models. It analytically computes moments for a Rauch–Tung–Striebel backward pass, accounting for uncertainty in the state and learned functions. Its output is a Gaussian approximation, not an exact posterior. In a simulated pendulum task, expected NLLx was 1.04 versus 10.3 for GP-URTSS, evidence limited to this setup.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.