Reproduce “Nonlinear Dynamical Modeling of Human Intracranial Brain Activity with Flexible Inference” 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

Kiarash Vaziri, Lucine L. Oganesian, HyeongChan Jo, Roberto M.C. Vera, Charles Y. Liu, Brian Lee, Maryam M. Shanechi

DFINE forecasts power features from multisite human intracranial EEG using nonlinear manifold embeddings and linear latent dynamics. Kalman filtering propagates the latent state when observations are missing. Across 10 patients, DFINE beat a gated recurrent unit in 8 subjects and outperformed a linear state-space model, but its advantage over that model was not significant at 20% observed data in simulated dropouts.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.