Litcius/Paper detail

Regularized by Physics: Graph Neural Network Parametrized Potentials for the Description of Intermolecular Interactions

Moritz Thürlemann, Lennard Böselt, Sereina Riniker

2023Journal of Chemical Theory and Computation28 citationsDOIOpen Access PDF

Abstract

potential-energy surfaces. Given only information about the constituting elements, the molecular topology, and reference potential energies, the models successfully learn to assign atom types and corresponding FF parameters from scratch. The resulting models and parameters are validated on a wide range of experimentally and computationally derived properties of systems including dimers, pure liquids, and molecular crystals.

Topics & Concepts

Computer scienceParametrization (atmospheric modeling)Ab initioInterpretabilityRobustness (evolution)Topology (electrical circuits)Gradient descentArtificial neural networkStatistical physicsPhysicsArtificial intelligenceChemistryMathematicsQuantum mechanicsGeneCombinatoricsRadiative transferBiochemistryMachine Learning in Materials ScienceComputational Drug Discovery MethodsProtein Structure and Dynamics
Regularized by Physics: Graph Neural Network Parametrized Potentials for the Description of Intermolecular Interactions | Litcius