Reproduce “Learning Behaviorally Relevant Dynamics in Multimodal Neural Recordings” 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
Parastoo Azizeddin, Eray Erturk, Maryam M. Shanechi
BREM-NET combines spike and LFP recordings to predict behavior while representing neural activity not explained by that prediction. Its staged model trains behaviorally relevant latents to predict behavior, then residual latents to account for remaining neural signals. In a macaque reaching dataset, it improved behavior decoding and neural prediction over single-modality models, but evidence is limited to this task.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.