De novo exploration and self-guided learning of potential-energy surfaces

Abstract

Interatomic potential models based on machine learning (ML) are rapidly developing as tools for materials simulations. However, because of their flexibility, they require large fitting databases that are normally created with substantial manual selection and tuning of reference configurations. Here, we show that ML potentials can be built in a largely automated fashion, exploring and fitting potential-energy surfaces from the beginning (de novo) within one and the same protocol. The key enabling step is the use of a configuration-averaged kernel metric that allows one to select the few most relevant structures at each step. The resulting potentials are accurate and robust for the wide range of configurations that occur during structure searching, despite only requiring a relatively small number of single-point DFT calculations on small unit cells. We apply the method to materials with diverse chemical nature and coordination environments, marking a milestone toward the more routine application of ML potentials in physics, chemistry, and materials science.

References in corpus (13)

• Restoring the density-gradient expansion for exchange in solids and surfaces
• Generalized gradient approximation for solids and their surfaces
• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• Machine Learning Unifies the Modelling of Materials and Molecules
• High-pressure phases of silane
• On-the-fly machine learning force field generation: Application to melting points
• Machine-learning based interatomic potential for amorphous carbon
• Phase transitions of hybrid perovskites simulated by machine-learning force fields trained on-the-fly with Bayesian inference
• Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation
• Stratified construction of neural network based interatomic models for multicomponent materials
• Conceptual and practical bases for the high accuracy of machine learning interatomic potential
• Revealing and exploiting hierarchical material structure through complex atomic networks
• Geometries of edge and mixed dislocations in bcc Fe from first principles calculations

Cited by in corpus (44)

• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• Big-Data Science in Porous Materials: Materials Genomics and Machine Learning
• Physics-inspired structural representations for molecules and materials
• Machine learning for electronically excited states of molecules
• Machine-learning interatomic potentials for materials science
• Towards high-throughput superconductor discovery via machine learning
• Machine Learning Inter-Atomic Potentials Generation Driven by Active Learning: A Case Study for Amorphous and Liquid Hafnium dioxide
• Active learning of reactive Bayesian force fields: Application to heterogeneous hydrogen-platinum catalysis dynamics
• How to validate machine-learned interatomic potentials
• Deep Potential generation scheme and simulation protocol for the Li10GeP2S12-type superionic conductors
• A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry
• Automated discovery of a robust interatomic potential for aluminum
• Strategies for the Construction of Machine-Learning Potentials for Accurate and Efficient Atomic-Scale Simulations
• Gaussian approximation potentials for body-centered-cubic transition metals
• Sparse Gaussian Process Potentials: Application to Lithium Diffusivity in Superionic Conducting Solid Electrolytes
• 2022 Review of Data-Driven Plasma Science
• Ephemeral data derived potentials for random structure search
• MAISE: Construction of neural network interatomic models and evolutionary structure optimization
• Atomic Permutationally Invariant Polynomials for Fitting Molecular Force Fields
• Systematic Atomic Structure Datasets for Machine Learning Potentials: Application to Defects in Magnesium
• Lifelong Machine Learning Potentials
• On Machine Learning Force Fields for Metallic Nanoparticles
• Combining phonon accuracy with high transferability in Gaussian approximation potential models
• Atomistic Structure Learning Algorithm with surrogate energy model relaxation
• Exploring the robust extrapolation of high-dimensional machine learning potentials
• Metadynamics sampling in atomic environment space for collecting training data for machine learning potentials
• Model-free quantification of completeness, uncertainties, and outliers in atomistic machine learning using information theory
• Multi-Fidelity Machine Learning for Excited State Energies of Molecules
• Atomistic structure search using local surrogate mode
• Accelerating Cathode Material Discovery through Ab Initio Random Structure Searching
• Machine Learned Interatomic Potentials for Ternary Carbides trained on the AFLOW Database
• A machine learning potential for simulating infrared spectra of nanosilicate clusters
• Optimized Multifidelity Machine Learning for Quantum Chemistry
• Mean-Field Density Matrix Decompositions
• Accelerated calculation of configurational free energy using a combination of reverse Monte Carlo and neural network models: Adsorption isotherm for 2D square and triangular lattices
• Machine Learning Enhanced Calculation of Quantum-Classical Binding Free Energies
• Accelerating the theoretical study of Li-polysulphide adsorption on single-atom catalysts via machine learning approaches
• Machine learning based modeling of disordered elemental semiconductors: understanding the atomic structure of a-Si and a-C
• Active-learning-based efficient prediction of ab-initio atomic energy: a case study on a Fe random grain boundary model with millions of atoms
• Scalable Data-Driven Basis Selection for Linear Machine Learning Interatomic Potentials
• Active Δ-learning with universal potentials for global structure optimization
• Compressing local atomic neighbourhood descriptors
• Predictive Synthesis of Quantum Materials by Probabilistic Reinforcement Learning
• Active learning potentials for first-principles phase diagrams using replica-exchange nested sampling