Constraint-aware neural networks clone known semilocal XC functionals more accurately in self-consistent calculations, transfer well from molecules to solids, and outperform unconstrained models across multiple tests.
{xcquinox}: A machine learning package using the Equinox library for learning exchange–correlation func- tionals with JAX
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.chem-ph 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Constraint-aware functional cloning for stable and transferable machine-learned density functional theory
Constraint-aware neural networks clone known semilocal XC functionals more accurately in self-consistent calculations, transfer well from molecules to solids, and outperform unconstrained models across multiple tests.