Reproduce “SymbolFit: Automatic Parametric Modeling with Symbolic Regression” 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
Ho Fung Tsoi, Dylan Rankin, Cecile Caillol, Miles Cranmer, Sridhara Dasu, Javier Duarte, Philip Harris, Elliot Lipeles and 1 more
SymbolFit, a framework developed by a collaboration including the University of Pennsylvania and CERN, automates parametric modeling of scientific data using symbolic regression combined with parameter re-optimization. It discovers functional forms and provides robust uncertainty estimates, performing comparably to manually derived empirical models in high-energy physics applications while significantly reducing human effort.
Key insights
- Symbolic regression, combined with robust re-optimization, can automate functional form discovery and provide crucial parameter uncertainty estimates for scientific data.
- The ability to generate multiple candidate functions offers flexibility for model selection and facilitates systematic uncertainty treatment.
- The framework directly addresses the needs of high-energy physics by producing closed-form functions with well-defined uncertainties, compatible with existing inference tools.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.