Reproduce “Symbolic Density Estimators for Unnormalized Distributions” 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
Vikas Kanaujia, Riyansha Singh, Shashank Sharma, Vipul Arora
The authors present a framework that turns samples from physical systems into symbolic energy laws by using learned density or score information to guide symbolic regression. For lattice systems, it searches local patches and adds noise to improve coverage. In non-perturbative ϕ⁴ renormalization, estimated Hamiltonians reproduce measured observable distributions, but this does not establish the exact or unique coarse-scale law.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.