Automated Construction of Neural Network Potential Energy Surface: The Enhanced Self-Organizing Incremental Neural Network Deep Potential Method

In recent years, the use of deep learning (neural network) potential energy surface (NNPES) in molecular dynamics simulation has experienced explosive growth as it can be as accurate as quantum chemistry methods while being as efficient as classical mechanic methods. However, the development of NNPES is highly nontrivial. In particular, it has been troubling to construct a dataset that is as small as possible yet can cover the target chemical space. In this work, an ESOINN-DP method is developed, which has the enhanced self-organizing incremental neural network (ESOINN) and a newly proposed error indicator at its core. With ESOINN-DP, one can construct the NNPES with little human intervention, and this method ensures that the constructed reference dataset covers the target chemical space with minimum redundancy. The performance of the ESOINN-DP method has been well validated by developing neural network potential energy surfaces for water clusters, tripeptides, and by de-redundancy of a sub-dataset of the ANI-1 database. We believe that the ESOINN-DP method provides a novel idea for the construction of NNPES and, especially, the reference datasets, and it can be used for molecular dynamics (MD) simulations of various gas-phase and condensed-phase chemical systems.

• Publication year

  2021

• Venue

  Journal of Chemical Information and Modeling

• Publication date

  2021-04-06

• Fields of study

  Chemistry, Medicine, Computer Science

• Identifiers
  DOI 10.26434/CHEMRXIV.14370527.V1PMID 34752095
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• Machine Learning for Chemical Reactions.

  2021cited by this paper
• Active learning accelerates ab initio molecular dynamics on reactive energy surfaces

  2021cited by this paper
• Searching Configurations in Uncertainty Space: Active Learning of High-Dimensional Neural Network Reactive Potentials.

  2021cited by this paper
• 86 PFLOPS Deep Potential Molecular Dynamics simulation of 100 million atoms with ab initio accuracy

  2020cited by this paper
• A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer

  2020cited by this paper
• Automated Construction of Potential Energy Surfaces Suitable to Describe van der Waals Complexes With Highly-Excited Nascent Molecules: The Rotational Spectra of Ar-CS(v) and Ar-SiS(v).

  2020cited by this paper
• A Bayesian Machine Learning Approach to the Quantification of Uncertainties on Ab Initio Potential Energy Surfaces.

  2020cited by this paper
• High-Fidelity Potential Energy Surfaces for Gas Phase and Gas-Surface Scattering Processes from Machine Learning.

  2020cited by this paper
• Quantum machine learning using atom-in-molecule-based fragments selected on the fly

  2020cited by this paper
• Active learning and neural network potentials accelerate molecular screening of ether-based solvate ionic liquids.

  2020cited by this paper
• New Perspectives on CO2–Pt(111) Interaction with a High-Dimensional Neural Network Potential Energy Surface

  2020cited by this paper
• Automatically growing global reactive neural network potential energy surfaces: A trajectory-free active learning strategy.

  2020cited by this paper
• Dynamics of H2O Adsorption on Pt(110)-(1 × 2) Based on a Neural Network Potential Energy Surface

  2020cited by this paper
• Neural Network Potential Energy Surfaces for Small Molecules and Reactions.

  2020cited by this paper
• Molecular force fields with gradient-domain machine learning: Construction and application to dynamics of small molecules with coupled cluster forces.

  2019cited by this paper
• High-dimensional potential energy surfaces for molecular simulations: from empiricism to machine learning

  2019cited by this paper
• Embedded Atom Neural Network Potentials: Efficient and Accurate Machine Learning with a Physically Inspired Representation.

  2019cited by this paper
• PES-Learn: An Open-Source Software Package for the Automated Generation of Machine Learning Models of Molecular Potential Energy Surfaces.

  2019cited by this paper
• Bio-inspired synthesis of xishacorenes A, B, and C, and a new congener from fuscol

  2019cited by this paper
• DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models

  2019cited by this paper
• Parallel Multistream Training of High-Dimensional Neural Network Potentials.

  2019cited by this paper
• Theoretical Investigations of Rate Coefficients of H + H2O2 → OH + H2O on a Full-Dimensional Potential Energy Surface.

  2019cited by this paper
• Incorporating long-range physics in atomic-scale machine learning.

  2019cited by this paper
• Discovery of Descriptors for Stable Monolayer Oxide Coatings through Machine Learning

  2018cited by this paper
• Permutationally Invariant Potential Energy Surfaces.

  2018cited by this paper
• Force Field for Water Based on Neural Network.

  2018cited by this paper
• Atomic Energies from a Convolutional Neural Network.

  2018cited by this paper
• Automatic selection of atomic fingerprints and reference configurations for machine-learning potentials.

  2018cited by this paper
• Comparison of permutationally invariant polynomials, neural networks, and Gaussian approximation potentials in representing water interactions through many-body expansions.

  2018cited by this paper
• Toward Building Protein Force Fields by Residue-Based Systematic Molecular Fragmentation and Neural Network.

  2018cited by this paper
• AUTOSURF: A Freely Available Program To Construct Potential Energy Surfaces

  2018cited by this paper
• Less is more: sampling chemical space with active learning

  2018cited by this paper
• Machine-Learning-Assisted Accurate Band Gap Predictions of Functionalized MXene

  2018cited by this paper
• The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics

  2017cited by this paper
• First Principles Neural Network Potentials for Reactive Simulations of Large Molecular and Condensed Systems.

  2017cited by this paper
• Multi-fidelity machine learning models for accurate bandgap predictions of solids

  2017cited by this paper
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics

  2017cited by this paper
• ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost

  2016cited by this paper
• OpenMM 7: Rapid development of high performance algorithms for molecular dynamics

  2016cited by this paper
• Perspective: Machine learning potentials for atomistic simulations.

  2016cited by this paper
• Quantum-chemical insights from deep tensor neural networks

  2016cited by this paper
• Machine learning of accurate energy-conserving molecular force fields

  2016cited by this paper
• Machine Learning Force Fields: Construction, Validation, and Outlook

  2016cited by this paper
• Genetic Optimization of Training Sets for Improved Machine Learning Models of Molecular Properties.

  2016cited by this paper
• Active learning of linearly parametrized interatomic potentials

  2016cited by this paper
• Molecular dynamics with on-the-fly machine learning of quantum-mechanical forces.

  2015cited by this paper
• Constructing high‐dimensional neural network potentials: A tutorial review

  2015cited by this paper
• Representing the potential-energy surface of protonated water clusters by high-dimensional neural network potentials.

  2015cited by this paper
• Solution coating around ice particles of incipient cirrus clouds

  2013cited by this paper
• Assessment and Validation of Machine Learning Methods for Predicting Molecular Atomization Energies.

  2013cited by this paper
• A density-functional theory-based neural network potential for water clusters including van der Waals corrections.

  2013cited by this paper
• A Full-Dimensional Neural Network Potential-Energy Surface for Water Clusters up to the Hexamer

  2013cited by this paper
• I and J

  2012cited by this paper
• Fast and accurate modeling of molecular atomization energies with machine learning.

  2011cited by this paper
• Atom-centered symmetry functions for constructing high-dimensional neural network potentials.

  2011cited by this paper
• Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations.

  2011cited by this paper
• Benchmark structures and binding energies of small water clusters with anharmonicity corrections.

  2011cited by this paper
• Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons.

  2009cited by this paper
• DFTB+, a sparse matrix-based implementation of the DFTB method.

  2007cited by this paper
• Generalized neural-network representation of high-dimensional potential-energy surfaces.

  2007cited by this paper
• Representing molecule-surface interactions with symmetry-adapted neural networks.

  2007cited by this paper
• An enhanced self-organizing incremental neural network for online unsupervised learning

  2007cited by this paper
• Ab initio potential-energy surfaces for complex, multichannel systems using modified novelty sampling and feedforward neural networks.

  2005cited by this paper
• Molecular potential energy surfaces by interpolation: strategies for faster convergence.

  2004cited by this paper
• Interpolating moving least-squares methods for fitting potential energy surfaces: Illustrative approaches and applications

  2003cited by this paper
• Molecular potential-energy surfaces for chemical reaction dynamics

  2002cited by this paper
• POLYATOMIC MOLECULAR POTENTIAL ENERGY SURFACES BY INTERPOLATION IN LOCAL INTERNAL COORDINATES

  1998cited by this paper
• Convergence of molecular potential energy surfaces by interpolation: Application to the OH+H2→H2O+H reaction

  1995cited by this paper
• Neural network models of potential energy surfaces

  1995cited by this paper
• AND T

  year unknowncited by this paper

• Autonomous Reaction Discovery of CO2 Capture in Aqueous Ammonia through Active-Learning Neural Networks.

  2026cites this paper
• Nano-key technology of semiconductor wireless signal recognition and processing based on self-organizing neural network

  2026cites this paper
• Transition State Searching Accelerated by Neural Network Potential

  2025cites this paper
• Minimizing Redundancy and Data Requirements of Machine Learning Potential: A Case Study in Interface Combustion.

  2024cites this paper
• From the Automated Calculation of Potential Energy Surfaces to Accurate Infrared Spectra

  2024cites this paper
• Rational Design of Earth‐Abundant Catalysts toward Sustainability

  2024cites this paper
• Electronic structure simulations in the cloud computing environment.

  2024cites this paper
• Active sampling for neural network potentials: Accelerated simulations of shear-induced deformation in Cu-Ni multilayers.

  2023cites this paper
• Digital twin modeling

  2022cites this paper
• 3D Conformational Generative Models for Biological Structures Using Graph Information-Embedded Relative Coordinates

  2022cites this paper
• Systematic Identification of Atom-Centered Symmetry Functions for the Development of Neural Network Potentials.

  2022cites this paper
• Automatically Constructed Neural Network Potentials for Molecular Dynamics Simulation of Zinc Proteins

  2021cites this paper
• Transition State Searching Accelerated by Deep Learning Potential

  year unknowncites this paper