Reproduce “Machine learning for lattice QCD” 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

Gurtej Kanwar, Scott Lawrence, Alessandro Nada, Phiala Shanahan

The authors review how machine learning can reduce lattice QCD computational costs while preserving correctness. They describe learned sampling, variance reduction, and surrogate observables, with corrections or mathematical identities retaining asymptotic correctness or exact expectations. A few full-scale surrogate studies saved about 20% of computational effort, insufficient to justify human effort; none was a default at scale by late 2025.

Summary by alphaXiv.

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