Reproduce “A latent control model for realistic rodent motion” 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
Aidan Sirbu, Charles Y. Zhang, Yuanjia Yang, Bence P. Ölveczky, Talmo D. Pereira, Blake A. Richards
SCAMPER adapts PULSE’s latent-controller approach to a simulated rat by combining a goal-directed motion prior with a behavior prior for varied exploration. A task policy selects between them, using residual control for exploitation and behavior-prior sampling for exploration. In simulated foraging, only full SCAMPER solves the largest minimum-radius conditions; this result does not establish animal-like learning.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.