Reproduce “Reinforcement Learning in High-Energy Physics: A Survey from Accelerator Control to Theory Exploration” 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
Luca Anzalone
The authors survey reinforcement learning in high-energy physics, organizing applications around control, monitoring, and sequential construction. They explain how policies learn actions from observations and rewards, and assess comparisons with traditional methods. In ARES accelerator tuning, the learned optimizer achieved median final beam error of 4 μm versus 45 μm for Bayesian optimization in simulation; this task-specific result is not a general verdict.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.