Reproduce “Sample-Efficient Reinforcement Learning Controller for Deep Brain Stimulation in Parkinson's Disease” 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
Harsh Ravivarapu, Gaurav Bagwe, Xiaoyong Yuan, Chunxiu Yu, Lan Zhang
SEA-DBS combines reward prediction and Gumbel-Softmax exploration to train a reinforcement-learning controller for adaptive deep brain stimulation in a simulated Parkinson’s disease model. It builds on DDPG and targets beta activity. At 75 steps, with environment seeds changing every n steps, GPi beta-band PSD was 302 versus 328 for DDPG. These results are simulation evidence, not patient validation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.