MGNN: Moment Graph Neural Network for Universal Molecular Potentials

Abstract

The quest for efficient and robust deep learning models for molecular systems representation is increasingly critical in scientific exploration. The advent of message passing neural networks has marked a transformative era in graph-based learning, particularly in the realm of predicting chemical properties and expediting molecular dynamics studies. We present the Moment Graph Neural Network (MGNN), a rotation-invariant message passing neural network architecture that capitalizes on the moment representation learning of 3D molecular graphs, is adept at capturing the nuanced spatial relationships inherent in three-dimensional molecular structures. MGNN demonstrates new state-of-the-art performance over contemporary methods on benchmark datasets such as QM9 and the revised MD17. The prowess of MGNN also extends to dynamic simulations, accurately predicting the structural and kinetic properties of complex systems such as amorphous electrolytes, with results that closely align with those from ab-initio simulations. The application of MGNN to the simulation of molecular spectra exemplifies its potential to significantly enhance the computational workflow, offering a promising alternative to traditional electronic structure methods

27 pages, 4 figures

References in corpus (29)

• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• SchNet - a deep learning architecture for molecules and materials
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Moment Tensor Potentials: a class of systematically improvable interatomic potentials
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges
• Assessing the Performance of Recent Density Functionals for Bulk Solids
• A Universal Graph Deep Learning Interatomic Potential for the Periodic Table
• By-passing the Kohn-Sham equations with machine learning
• Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields
• Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds
• Towards Universal Neural Network Potential for Material Discovery Applicable to Arbitrary Combination of 45 Elements
• The Phase Diagram of a Deep Potential Water Model
• sGDML: Constructing Accurate and Data Efficient Molecular Force Fields Using Machine Learning
• 86 PFLOPS Deep Potential Molecular Dynamics simulation of 100 million atoms with ab initio accuracy
• Accurate Interatomic Force Fields via Machine Learning with Covariant Kernels
• Forces are not Enough: Benchmark and Critical Evaluation for Machine Learning Force Fields with Molecular Simulations
• Transferable atomic multipole machine learning models for small organic molecules
• Unified Graph Neural Network Force-field for the Periodic Table
• Informing Geometric Deep Learning with Electronic Interactions to Accelerate Quantum Chemistry
• SchNetPack 2.0: A neural network toolbox for atomistic machine learning
• Extending the limit of molecular dynamics with ab initio accuracy to 10 billion atoms
• Spherical Channels for Modeling Atomic Interactions
• ænet-PyTorch: a GPU-supported implementation for machine learning atomic potentials training
• Smooth, exact rotational symmetrization for deep learning on point clouds

Cited by in corpus (1)

• Machine-Learned Potentials for Solvation Modeling