Developments and Further Applications of Ephemeral Data Derived Potentials

Abstract

Machine-learned interatomic potentials are fast becoming an indispensable tool in computational materials science. One approach is the ephemeral data-derived potential (EDDP), which was designed to accelerate atomistic structure prediction. The EDDP is simple and cost-efficient. It relies on training data generated in small unit cells and is fit using a lightweight neural network, leading to smooth interactions which exhibit the robust transferability essential for structure prediction. Here, we present a variety of applications of EDDPs, enabled by recent developments of the open-source EDDP software. New features include interfaces to phonon and molecular dynamics codes, as well as deployment of the ensemble deviation for estimating the confidence in EDDP predictions. Through case studies ranging from elemental carbon and lead to the binary scandium hydride and the ternary zinc cyanide, we demonstrate that EDDPs can be trained to cover wide ranges of pressures and stoichiometries, and used to evaluate phonons, phase diagrams, superionicity, and thermal expansion. These developments complement continued success in accelerated structure prediction.

arXiv comment22 pages, 15 figures

References in corpus (55)

• Restoring the density-gradient expansion for exchange in solids and surfaces
• Generalized gradient approximation for solids and their surfaces
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• Global Optimization by Basin-Hopping and the Lowest Energy Structures of Lennard-Jones Clusters Containing up to 110 Atoms
• Crystal Structure Prediction via Particle Swarm Optimization
• On representing chemical environments
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• Superconductivity at 250 K in lanthanum hydride under high pressures
• Moment Tensor Potentials: a class of systematically improvable interatomic potentials
• Evidence for superconductivity above 260 K in lanthanum superhydride at megabar pressures
• Ab initio Random Structure Searching
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• A Universal Graph Deep Learning Interatomic Potential for the Periodic Table
• Less is more: sampling chemical space with active learning
• High-pressure phases of silane
• Active learning of linearly parametrized interatomic potentials
• Finding Density Functionals with Machine Learning
• Physics-inspired structural representations for molecules and materials
• Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation
• Machine learning a general purpose interatomic potential for silicon
• Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning
• Ab initio thermodynamics of liquid and solid water
• The Lennard-Jones potential: when (not) to use it
• MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields
• An Accurate and Transferable Machine Learning Potential for Carbon
• Superconducting hydrides under pressure
• A neural network interatomic potential for the phase change material GeTe
• Data-driven learning of total and local energies in elemental boron
• Committee neural network potentials control generalization errors and enable active learning
• Atom-Density Representations for Machine Learning
• Evidence for supercritical behavior of high-pressure liquid hydrogen
• Efficient global structure optimization with a machine learned surrogate model
• Lattice dynamics and electron-phonon coupling calculations using non-diagonal supercells
• Feature Optimization for Atomistic Machine Learning Yields A Data-Driven Construction of the Periodic Table of the Elements
• Unified Graph Neural Network Force-field for the Periodic Table
• Atomic cluster expansion of scalar, vectorial and tensorial properties and including magnetism and charge transfer
• Accelerating CALYPSO Structure Prediction by Data-driven Learning of Potential Energy Surface
• Strategies for the Construction of Machine-Learning Potentials for Accurate and Efficient Atomic-Scale Simulations
• Stratified construction of neural network based interatomic models for multicomponent materials
• Search for ambient superconductivity in the Lu-N-H system
• Simulation study of pressure and temperature dependence of the negative thermal expansion in Zn(CN)${}_2$
• A Modern Take on the Bias-Variance Tradeoff in Neural Networks
• NOMAD 2018 Kaggle Competition: Solving Materials Science Challenges Through Crowd Sourcing
• PANNA: Properties from Artificial Neural Network Architectures
• MAISE: Construction of neural network interatomic models and evolutionary structure optimization
• First-principles study on the superconductivity of N-doped fcc-LuH3
• Systematic Atomic Structure Datasets for Machine Learning Potentials: Application to Defects in Magnesium
• Lu-H-N phase diagram from first-principles calculations
• Structure, Stability and Superconductivity of N-doped Lutetium Hydrides at kbar Pressures
• Global optimization of atomic structures with gradient-enhanced Gaussian process regression
• A first-principles study of co-doping in lanthanum bromide
• Microscopic theory of colour in lutetium hydride
• Lattice dynamics and elastic properties of alpha-U at high-temperature and high-pressure by machine learning potential simulations
• Ab initio machine learning of phase space averages
• Novel Porous Polymorphs of Zinc Cyanide with Rich Thermal and Mechanical Behavior

Cited by in corpus (10)

• Feasible route to high-temperature ambient-pressure hydride superconductivity
• The Evolution of Machine Learning Potentials for Molecules, Reactions and Materials
• Machine-learning structural reconstructions for accelerated point defect calculations
• Machine Learned Interatomic Potentials for Ternary Carbides trained on the AFLOW Database
• XtalOpt Version 13: Multi-Objective Evolutionary Search for Novel Functional Materials
• Application-specific machine-learned interatomic potentials: exploring the trade-off between DFT convergence, MLIP expressivity, and computational cost
• Guest Editorial: Special Topic on Software for Atomistic Machine Learning
• XtalOpt Version 14: Variable-Composition Crystal Structure Search for Functional Materials Through Pareto Optimization
• Iterative learning scheme for crystal structure prediction with anharmonic lattice dynamics
• Competing Hydrogenation Pathways to Metastable CaH${}_6$ Revealed by Machine-Learning-Potential Molecular Dynamics