Interpretable ML on limited Fe-catalyst data identifies thermodynamic lattice stability and geometric factors as primary drivers of electronic band gap over bulk stoichiometry, with non-linear models reaching R2 0.61-0.77 versus 0.32 for linear baselines.
Machine Learning on Contact Angles of Liquid Metals and Solid Oxides
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
-
Beyond the Black Box: An Interpretable Machine Learning Framework for Predicting Electronic Structure Microdescriptors and Structure-Performance Relationships in Fe-based Catalytic Systems
Interpretable ML on limited Fe-catalyst data identifies thermodynamic lattice stability and geometric factors as primary drivers of electronic band gap over bulk stoichiometry, with non-linear models reaching R2 0.61-0.77 versus 0.32 for linear baselines.