<i>CoeffNet</i> : predicting activation barriers through a chemically-interpretable, equivariant and physically constrained graph neural network
Sudarshan Vijay, Maxwell C. Venetos, Evan Walter Clark Spotte‐Smith, Aaron D. Kaplan, Mingjian Wen, Kristin A. Persson
Abstract
CoeffNet uses coefficients of molecular orbitals of reactants and products to predict activation barriers.
Topics & Concepts
Equivariant mapGraphAtomic orbitalArtificial neural networkComputational chemistryComputer scienceChemistryArtificial intelligenceTheoretical computer scienceMathematicsPhysicsPure mathematicsQuantum mechanicsElectronComputational Drug Discovery MethodsMachine Learning in Materials ScienceCO2 Reduction Techniques and Catalysts