Reproduce “RLOC: Neurobiologically Inspired Hierarchical Reinforcement Learning Algorithm for Continuous Control of Nonlinear 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

Ekaterina Abramova, Luke Dickens, Daniel Kuhn, Aldo Faisal

RLOC learns a global policy by combining local continuous controllers and learning when to choose each. It estimates local dynamics from experience and uses controller cost as feedback. In simulated cart-pole, with at least seven controllers, its average cost was lower than iLQR’s in 96% of trials and PILCO’s in 75%. Evidence is limited to two low-dimensional simulations.

Summary by alphaXiv.

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