High-dimensional potential energy surfaces for molecular simulations: from empiricism to machine learning
Oliver T Unke, Debasish Koner, Sarbani Patra, Silvan Käser, Markus Meuwly
Abstract
Abstract An overview of computational methods to describe high-dimensional potential energy surfaces suitable for atomistic simulations is given. Particular emphasis is put on accuracy, computability, transferability and extensibility of the methods discussed. They include empirical force fields, representations based on reproducing kernels, using permutationally invariant polynomials, neural network-learned representations and combinations thereof. Future directions and potential improvements are discussed primarily from a practical, application-oriented perspective.
Topics & Concepts
Artificial intelligenceComputer scienceTransferabilityMachine learningInvariant (physics)Energy (signal processing)Potential energyEmpiricismArtificial neural networkComputationComputational modelField (mathematics)Force field (fiction)Statistical physicsTheoretical computer scienceDeep neural networksPotential energy surfacePotential fieldEmpirical modellingFeature (linguistics)Machine Learning in Materials ScienceAdvanced Chemical Physics StudiesAdvanced Physical and Chemical Molecular Interactions