Embedded Atom Neural Network Potentials: Efficient and Accurate Machine Learning with a Physically Inspired Representation

Abstract

We propose a simple, but efficient and accurate machine learning (ML) model for developing high-dimensional potential energy surface. This so-called embedded atom neural network (EANN) approach is inspired by the well-known empirical embedded atom method (EAM) model used in condensed phase. It simply replaces the scalar embedded atom density in EAM with a Gaussian-type orbital based density vector, and represents the complex relationship between the embedded density vector and atomic energy by neural networks. We demonstrate that the EANN approach is equally accurate as several established ML models in representing both big molecular and extended periodic systems, yet with much fewer parameters and configurations. It is highly efficient as it implicitly contains the three-body information without an explicit sum of the conventional costly angular descriptors. With high accuracy and efficiency, EANN potentials can vastly accelerate molecular dynamics and spectroscopic simulations in complex systems at ab initio level.

The paper have been published on the journal of physical chemistry letters

References in corpus (17)

• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• 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
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• Machine learning a general purpose interatomic potential for silicon
• Efficient and Accurate Machine-Learning Interpolation of Atomic Energies in Compositions with Many Species
• Symmetry-Adapted Machine-Learning for Tensorial Properties of Atomistic Systems
• WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials
• Accurate molecular polarizabilities with coupled-cluster theory and machine learning
• Atom-Density Representations for Machine Learning
• A Universal Density Matrix Functional from Molecular Orbital-Based Machine Learning: Transferability across Organic Molecules
• Efficient non-parametric fitting of potential energy surfaces for polyatomic molecules with Gaussian processes
• Conceptual and practical bases for the high accuracy of machine learning interatomic potential

Cited by in corpus (45)

• Machine Learning Force Fields
• A Universal Graph Deep Learning Interatomic Potential for the Periodic Table
• A Performance and Cost Assessment of Machine Learning Interatomic Potentials
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• DeePMD-kit v2: A software package for Deep Potential models
• Neuroevolution machine learning potentials: Combining high accuracy and low cost in atomistic simulations and application to heat transport
• Machine learning for electronically excited states of molecules
• Neural Network Based in Silico Simulation of Combustion Reactions
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Neural Network Potentials for Chemistry: Concepts, Applications and Prospects
• DPA-2: a large atomic model as a multi-task learner
• TeaNet: universal neural network interatomic potential inspired by iterative electronic relaxations
• Gaussian Moments as Physically Inspired Molecular Descriptors for Accurate and Scalable Machine Learning Potentials
• Physically Motivated Recursively Embedded Atom Neural Networks: Incorporating Local Completeness and Nonlocality
• Efficient and Accurate Simulations of Vibrational and Electronic Spectra with Symmetry-Preserving Neural Network Models for Tensorial Properties
• Quantum deep field: data-driven wave function, electron density generation, and atomization energy prediction and extrapolation with machine learning
• Universal Machine Learning for the Response of Atomistic Systems to External Fields
• The Evolution of Machine Learning Potentials for Molecules, Reactions and Materials
• Symmetry-Adapted High Dimensional Neural Network Representation of Electronic Friction Tensor of Adsorbates on Metals
• A deep neural network for molecular wave functions in quasi-atomic minimal basis representation
• Predicting Properties of Periodic Systems from Cluster Data: A Case Study of Liquid Water
• Fast and Sample-Efficient Interatomic Neural Network Potentials for Molecules and Materials Based on Gaussian Moments
• Accelerating Atomistic Simulations with Piecewise Machine Learned Ab Initio Potentials at Classical Force Field-like Cost
• PyXtal FF: a Python Library for Automated Force Field Generation
• Mechanical Vibrational Relaxation of NO Scattering from Metal and Insulator Surfaces: When and Why Are They Different?
• An Extendible, Graph-Neural-Network-Based Approach for Accurate Force Field Development of Large Flexible Organic Molecules
• Machine Learning and Data-Driven Methods in Computational Surface and Interface Science
• Accelerating the prediction of inorganic surfaces with machine learning interatomic potentials
• Effect of surface temperature on quantum dynamics of H${}_2$ on Cu(111) using a chemically accurate potential energy surface
• How False Data Affects Machine Learning Models in Electrochemistry?
• Efficient force field and energy emulation through partition of permutationally equivalent atoms
• Extending the atomic decomposition and many-body representation, a chemistry-motivated monomer-centered approach for machine learning potentials
• A simple approach to rotationally invariant machine learning of avector quantity
• Reaction dynamics for the Cl(${}^2$P) + XCl $\to$ XCl + Cl(${}^2$P) (X = H, D, Mu) reaction on a high-fidelity ground state potential energy surface
• Ab-Initio Potential Energy Surfaces by Pairing GNNs with Neural Wave Functions
• Lattice ground states for Embedded-Atom Models in 2D and 3D
• On the equivalence of molecular graph convolution and molecular wave function with poor basis set
• Physical machine learning outperforms "human learning" in Quantum Chemistry
• Learning the Electrostatic Response of the Electron Density through a Symmetry-Adapted Vector Field Model
• FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs
• Geometric learning of the conformational dynamics of molecules using dynamic graph neural networks
• Active Learning of a Neural Network Potential for Large-Scale Atomistic Simulations of Polymer Electrolyte Membranes
• REANN: A PyTorch-based End-to-End Multi-functional Deep Neural Network Package for Molecular, Reactive and Periodic Systems
• Detect the Interactions that Matter in Matter: Geometric Attention for Many-Body Systems
• Neural Polarization: Toward Electron Density for Molecules by Extending Equivariant Networks