Active learning of linearly parametrized interatomic potentials

Abstract

This paper introduces an active learning approach to the fitting of machine learning interatomic potentials. Our approach is based on the D-optimality criterion for selecting atomic configurations on which the potential is fitted. It is shown that the proposed active learning approach is highly efficient in training potentials on the fly, ensuring that no extrapolation is attempted and leading to a completely reliable atomistic simulation without any significant decrease in accuracy. We apply our approach to molecular dynamics and structure relaxation, and we argue that it can be applied, in principle, to any other type of atomistic simulation. The software, test cases, and examples of usage are published at http://gitlab.skoltech.ru/shapeev/mlip/.

12 pages, numerical tests added

References in corpus (5)

• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• Machine-learning based interatomic potential for amorphous carbon
• Accurate Interatomic Force Fields via Machine Learning with Covariant Kernels
• Machine learning force fields: Construction, validation, and outlook

Cited by in corpus (152)

• SchNet - a deep learning architecture for molecules and materials
• DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models
• A Performance and Cost Assessment of Machine Learning Interatomic Potentials
• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• Less is more: sampling chemical space with active learning
• Towards Exact Molecular Dynamics Simulations with Machine-Learned Force Fields
• On-the-fly machine learning force field generation: Application to melting points
• Big-Data Science in Porous Materials: Materials Genomics and Machine Learning
• Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation
• Machine learning a general purpose interatomic potential for silicon
• GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations
• SchNetPack: A Deep Learning Toolbox For Atomistic Systems
• Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning
• Machine learning for electronically excited states of molecules
• Machine-learning interatomic potentials for materials science
• Autonomous discovery in the chemical sciences part I: Progress
• Hierarchical modeling of molecular energies using a deep neural network
• sGDML: Constructing Accurate and Data Efficient Molecular Force Fields Using Machine Learning
• De novo exploration and self-guided learning of potential-energy surfaces
• Accelerating first-principles estimation of thermal conductivity by machine-learning interatomic potentials: A MTP/ShengBTE solution
• Machine-Learning Interatomic Potentials Enable First-Principles Multiscale Modeling of Lattice Thermal Conductivity in Graphene/Borophene Heterostructures
• Data-driven learning of total and local energies in elemental boron
• Extending the Accuracy of the SNAP Interatomic Potential Form
• Deep Potentials for Materials Science
• A practical guide to machine learning interatomic potentials -- Status and future
• Committee neural network potentials control generalization errors and enable active learning
• Efficient global structure optimization with a machine learned surrogate model
• Machine learning of molecular properties: locality and active learning
• Nanoporous C3N4, C3N5 and C3N6 nanosheets; Novel strong semiconductors with low thermal conductivities and appealing optical/electronic properties
• Exploring Phononic Properties of Two-Dimensional Materials using Machine Learning Interatomic Potentials
• Representations of molecules and materials for interpolation of quantum-mechanical simulations via machine learning
• Efficient nonparametric $n$-body force fields from machine learning
• Deep neural network for the dielectric response of insulators
• Active learning of reactive Bayesian force fields: Application to heterogeneous hydrogen-platinum catalysis dynamics
• Automated Fitting of Neural Network Potentials at Coupled Cluster Accuracy: Protonated Water Clusters as Testing Ground
• How to validate machine-learned interatomic potentials
• Ab initio vibrational free energies including anharmonicity for multicomponent alloys
• A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry
• Molecular Force Fields with Gradient-Domain Machine Learning: Construction and Application to Dynamics of Small Molecules with Coupled Cluster Forces
• Machine Learning for Predicting Thermal Transport Properties of Solids
• Automated discovery of a robust interatomic potential for aluminum
• A local Bayesian optimizer for atomic structures
• Gaussian approximation potential modeling of lithium intercalation in carbon nanostructures
• Strategies for the Construction of Machine-Learning Potentials for Accurate and Efficient Atomic-Scale Simulations
• Atomic-scale origin of the low grain-boundary resistance in perovskite solid electrolytes
• Tutorial: How to Train a Neural Network Potential
• Modelling atomic and nanoscale structure in the silicon-oxygen system through active machine learning
• Global optimization of atomistic structure enhanced by machine learning
• Fast Uncertainty Estimates in Deep Learning Interatomic Potentials
• Structural, electronic, thermal and mechanical properties of C60-based fullerene two-dimensional networks explored by first-principles and machine learning
• Data-driven Material Models for Atomistic Simulation
• Effect of paramagnetic impurities on superconductivity in polyhydrides: $\text{s}$-wave order parameter in Nd-doped LaH${}_{10}$
• Exploring thermal expansion of carbon-based nanosheets by machine-learning interatomic potentials
• Novel Strongly Correlated Europium Superhydrides
• The Resistive Nature of Decomposing Interfaces of Solid Electrolytes with Alkali Metal Electrodes
• Bayesian Force Fields from Active Learning for Simulation of Inter-Dimensional Transformation of Stanene
• High-temperature superconductivity in hydrides
• Discrepancies and Error Evaluation Metrics for Machine Learning Interatomic Potentials
• An Experimentally Driven Automated Machine Learned lnter-Atomic Potential for a Refractory Oxide
• MAISE: Construction of neural network interatomic models and evolutionary structure optimization
• Molecular Force Fields with Gradient-Domain Machine Learning (GDML): Comparison and Synergies with Classical Force Fields
• Challenges for Machine Learning Force Fields in Reproducing Potential Energy Surfaces of Flexible Molecules
• Uncertainty-biased molecular dynamics for learning uniformly accurate interatomic potentials
• Dynamical Strengthening of Covalent and Non-Covalent Molecular Interactions by Nuclear Quantum Effects at Finite Temperature
• Systematic Atomic Structure Datasets for Machine Learning Potentials: Application to Defects in Magnesium
• Optimization and Validation of a Deep Learning CuZr Atomistic Potential: Robust Applications for Crystalline and Amorphous Phases with near-DFT Accuracy
• Accurate machine learning force fields via experimental and simulation data fusion
• Automated Calculation of Thermal Rate Coefficients using Ring Polymer Molecular Dynamics and Machine-Learning Interatomic Potentials with Active Learning
• Transfer learning for chemically accurate interatomic neural network potentials
• Lifelong Machine Learning Potentials
• Developments and Further Applications of Ephemeral Data Derived Potentials
• Extending the limit of molecular dynamics with ab initio accuracy to 10 billion atoms
• Fast and Sample-Efficient Interatomic Neural Network Potentials for Molecules and Materials Based on Gaussian Moments
• Elinvar effect in $\beta -$Ti simulated by on-the-fly trained moment tensor potential
• Thermo-mechanical properties of nitrogenated holey graphene (C2N): A comparison of machine-learning-based and classical interatomic potentials
• Thermophysical properties of FLiBe using moment tensor potentials
• Exploring the robust extrapolation of high-dimensional machine learning potentials
• Atomistic Structure Learning Algorithm with surrogate energy model relaxation
• Layer-by-layer phase transformation in Ti${}_3$O${}_5$ revealed by machine learning molecular dynamics simulations
• Machine-learning potentials for nanoscale simulations of deformation and fracture: example of TiB${}_2$ ceramic
• Sr-Doped Molecular Hydrogen: Synthesis and Properties of SrH${}_{22}$
• Accurate estimation of interfacial thermal conductance between silicon and diamond enabled by a machine learning interatomic potential
• Integrating Machine Learning with Physics-Based Modeling
• A deep machine learning potential for atomistic simulation of Fe-Si-O systems under Earth's outer core conditions
• Machine-learning potentials enable predictive $\text{and}$ tractable high-throughput screening of random alloys
• Indirect Learning of Interatomic Potentials for Accelerated Materials Simulations
• Machine Learned Interatomic Potential for Dispersion Strengthened Plasma Facing Components
• High-dimensional neural network potentials for accurate vibrational frequencies: The formic acid dimer benchmark
• Reliable emulation of complex functionals by active learning with error control
• In operando active learning of interatomic interaction during large-scale simulations
• Active learning of effective Hamiltonian for super-large-scale atomic structures
• Machine Learned Interatomic Potentials for Ternary Carbides trained on the AFLOW Database
• Constructing and evaluating machine-learned interatomic potentials for Li-based disordered rocksalts
• Machine Learning with bond information for local structure optimizations in surface science
• Modeling Extensive Defects in Metals through Classical Potential-Guided Sampling and Automated Configuration Reconstruction
• Ring Polymer Molecular Dynamics and Active Learning of Moment Tensor Potential for Gas-Phase Barrierless Reactions: Application to S + H2
• The fundamentals of quantum machine learning
• Lattice thermal conductivity and elastic modulus of XN4 (X=Be, Mg and Pt) 2D materials using machine learning interatomic potentials
• Machine-Learned Atomic Cluster Expansion Potentials for Fast and Quantum-Accurate Thermal Simulations of Wurtzite AlN
• Best Practices for Fitting Machine Learning Interatomic Potentials for Molten Salts: A Case Study Using NaCl-MgCl2
• Bayesian learning of thermodynamic integration and numerical convergence for accurate phase diagrams
• Ab initio Canonical Sampling based on Variational Inference
• Atomistic modeling of bulk and grain boundary diffusion in solid electrolyte Li${}_6$PS${}_5$Cl using machine-learning interatomic potentials
• Unsupervised learning-based structural analysis: Search for a characteristic low-dimensional space by local structures in atomistic simulations
• Multiscale Modeling of Metal/Oxide/Metal Conductive Bridging Random Access Memory Cells: from Ab Initio to Finite Element Calculations
• Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights
• Facilitating {\it ab initio} configurational sampling of multicomponent solids using an on-lattice neural network model and active learning
• Accurate Molecular Dynamics Enabled by Efficient Physically-Constrained Machine Learning Approaches
• Ab initio framework for deciphering trade-off relationships in multi-component alloys
• Machine learning potentials for complex aqueous systems made simple
• Machine Learning Potentials for Hydrogen Absorption in TiCr${}_2$ Laves Phases
• Fitting to magnetic forces improves the reliability of magnetic Moment Tensor Potentials
• Machine Learning Enhanced Calculation of Quantum-Classical Binding Free Energies
• Fast proper orthogonal descriptors for many-body interatomic potentials
• Automated atomistic simulations of dissociated dislocations with ab initio accuracy
• Efficient moment tensor machine-learning interatomic potential for accurate description of defects in Ni-Al Alloys
• AI-accelerated Materials Informatics Method for the Discovery of Ductile Alloys
• Efficient Generation of Stable Linear Machine-Learning Force Fields with Uncertainty-Aware Active Learning
• Enhanced sampling of robust molecular datasets with uncertainty-based collective variables
• Bayesian inference of composition-dependent phase diagrams
• Active-learning-based efficient prediction of ab-initio atomic energy: a case study on a Fe random grain boundary model with millions of atoms
• Apax: A Flexible and Performant Framework For The Development of Machine-Learned Interatomic Potentials
• Compressing and forecasting atomic material simulations with descriptors
• Machine learning interatomic potential for predicting the thermal properties of uranium nitride
• Accessing negative Poisson`s ratio of graphene by machine learning interatomic potentials
• Evolutionary Monte Carlo of QM properties in chemical space: Electrolyte design
• Application-specific machine-learned interatomic potentials: exploring the trade-off between DFT convergence, MLIP expressivity, and computational cost
• Ab initio machine learning in chemical compound space
• Machine-Learning Potentials Predict Orientation- and Mode-Dependent Fracture in Refractory Diborides
• Free energy profiles for chemical reactions in solution from high-dimensional neural network potentials: The case of the Strecker synthesis
• Machine-Learning Surrogate Model for Accelerating the Search of Stable Ternary Alloys
• Incipient ionic conductors: Ion-constrained lattices achieving superionic-like thermal conductivity by extreme anharmonicity
• Machine-learning interatomic potentials achieving CCSD(T) accuracy for systems with extended covalent networks and van der Waals interactions
• Efficient GPU-Accelerated Training of a Neuroevolution Potential with Analytical Gradients
• Low-rank matrix and tensor approximations for compression of machine-learning interatomic potentials
• Hydrogen diffusion in TiCr${}_2$H${}_x$ Laves phases: A combined ab initio and machine-learning-potential study
• On Simulating Thin-Film Processes at the Atomic Scale Using Machine Learned Force Fields
• An information-matching approach to optimal experimental design and active learning
• k-Means Clustering in Fingerprint-Based Configuration Selection for Fitting Interatomic Potentials
• Exact average many-body interatomic interaction model for random alloys
• Entropy-based Active Learning of Graph Neural Network Surrogate Models for Materials Properties
• Benchmarking Chemically Scalable Machine-Learning Interatomic Potentials for Large-Scale Simulations of Multicomponent Alloys
• Moment tensor Potentials as a Promising Tool to Study Diffusion Processes
• Evaluating approaches for on-the-fly machine learning interatomic potential for activated mechanisms sampling with the activation-relaxation technique nouveau
• Integer linear programming for unsupervised training set selection in molecular machine learning
• Active learning of potential-energy surfaces of weakly-bound complexes with regression-tree ensembles
• Thermal transport and phase transitions of zirconia by on-the-fly machine-learned interatomic potentials
• Atomistic mechanisms of phase transitions in all-temperature barocaloric material KPF${}_6$
• Analysis of Atomistic Representations Using Weighted Skip-Connections
• Machine-Learned Interatomic Potentials for Predicting Physicochemical Properties of Molten Metal-Salt Systems for Calcium Electrolysis
• Inverse design of bespoke interatomic potentials via active learning by information-matching
• Assessing zero-shot generalisation behaviour in graph-neural-network interatomic potentials