Reproduce “Identifying Neural Dynamics Using Nonlinear Interventional 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
Hanna M. Tolle, Carles Balsells-Rodas, Yixin Wang, Pedro A. M. Mediano, Amin Nejatbakhsh
The authors introduce a nonlinear interventional state-space model that uses stimulation to constrain neural dynamics. Under assumptions, interventions narrow latent-coordinate ambiguity to permutation, scaling, and shift. In intracranial EEG, its open-loop rollout accuracy at one stimulation site withheld from training was 0.71 ± 0.02, versus 0.37 ± 0.02 for a linear model; this is proof of concept, not a guarantee for unseen sites.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.