Reproduce “Electron-Density-Informed Molecular Property Prediction Using Surrogate Functionals” 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
Manuel V. Klockow, Simon Wagner, Roman Remme, Dominic Plein, Fred A. Hamprecht
The authors extend surrogate-functional training to predict scalar molecular properties using a learned property function whose minimum is the molecule’s ground-state electron density. On QM9 and selected QM40 molecules, the method lowers errors for extensive properties relative to density-input alternatives, but not for every property; the mechanism remains unknown.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.