Reproduce “Variational Gaussian Process State-Space Models” 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
Roger Frigola, Yutian Chen, Carl E. Rasmussen
The variational GP-SSM learns an approximate posterior over nonlinear dynamics. It represents the transition function with sparse Gaussian-process inducing variables and uses variational inference while smoothing hidden states. In a rat hippocampal example, its forecast captures a roughly 6 Hz pattern with process noise off. On a nonlinear test, PMCMC is slightly more accurate; reported times are shorter, but its training-code caveat limits timing.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.