Reproduce “LFADS - Latent Factor Analysis via Dynamical Systems” 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
David Sussillo, Rafal Jozefowicz, L. F. Abbott, Chethan Pandarinath
LFADS models neural activity as the output of a low-dimensional evolving dynamical system, while a controller can infer inputs when those dynamics do not explain observed spikes. In synthetic tests, it outperformed comparison methods on latent-variable recovery and held-out firing-rate estimates, and often located injected-pulse timing. These results concern simulated data, not validation on real neural recordings.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.