A Performance and Cost Assessment of Machine Learning Interatomic Potentials.

  title={A Performance and Cost Assessment of Machine Learning Interatomic Potentials.},
  author={Yunxing Zuo and Chi Chen and Xiang-Guo Li and Zhi Chang Deng and Yiming Chen and J{\"o}rg Behler and G{\'a}bor Cs{\'a}nyi and Alexander V. Shapeev and Aidan P. Thompson and Mitchell A. Wood and Shyue Ping Ong},
  journal={The journal of physical chemistry. A},
Machine learning of the quantitative relationship between local environment descriptors and the potential energy surface of a system of atoms has emerged as a new frontier in the development of interatomic potentials (IAPs). Here, we present a comprehensive evaluation of ML-IAPs based on four local environment descriptors --- atom-centered symmetry functions (ACSF), smooth overlap of atomic positions (SOAP), the Spectral Neighbor Analysis Potential (SNAP) bispectrum components, and moment… 

Figures and Tables from this paper

Dual adaptive sampling and machine learning interatomic potentials for modeling materials with chemical bond hierarchy
The development of reliable and flexible machine learning based interatomic potentials (ML-IPs) is becoming increasingly important in studying the physical properties of complex condensed matter
Development of a general-purpose machine-learning interatomic potential for aluminum by the physically informed neural network method
Abstract Interatomic potentials constitute the key component of large-scale atomistic simulations of materials. The recently proposed physically-informed neural network (PINN) method combines a
Training Data Selection for Accuracy and Transferability of Interatomic Potentials
This work demonstrates the development of both accurate and transferable ML potentials using automated and data-driven approaches for generating large and diverse training sets for tungsten in an automated manner.
Ultra-fast interpretable machine-learning potentials
Effective two and three-body potentials in a cubic B-spline basis with regularized linear regression are combined to obtain machine-learning potentials that are physically interpretable, sufficiently accurate for applications, as fast as the fastest traditional empirical potentials, and two to four orders of magnitude faster than state-of-the-art machine- learning potentials.
Bridging the Efficiency Gap Between Machine Learned Potentials with ab initio Accuracy and Classical Force Fields
An embedded atom neural network (EANN) approach with simple piecewise switching function based descriptors, resulting in a favorable linear scaling with the number of neighbor atoms, approaching the speed of fastest embedded atom method.
Machine-learning potentials enable predictive and tractable high-throughput screening of random alloys
An algorithm for training moment tensor potentials (MTPs) with the accuracy of density functional theory (DFT) and forms the basis for an efficient construction of MTPs to be used in large-scale simulations of multicomponent systems.
Machine-learned interatomic potentials by active learning: amorphous and liquid hafnium dioxide
An active learning scheme for automatically sampling a minimum number of uncorrelated configurations for fitting the Gaussian Approximation Potential (GAP) model is proposed and shown to be able to extract a configuration that reaches the required energy fit tolerance.


Accurate Force Field for Molybdenum by Machine Learning Large Materials Data
It is shown that by fitting to the energies, forces and stress tensors of a large density functional theory (DFT)-computed dataset on a diverse set of Mo structures, a Mo SNAP model can be developed that achieves close to DFT accuracy in the prediction of a broad range of properties.
Achieving DFT accuracy with a machine-learning interatomic potential: thermomechanics and defects in bcc ferromagnetic iron
We show that the Gaussian Approximation Potential machine learning framework can describe complex magnetic potential energy surfaces, taking ferromagnetic iron as a paradigmatic challenging case. The
Quantum-accurate spectral neighbor analysis potential models for Ni-Mo binary alloys and fcc metals
A systematic model development process for multicomponent alloy systems is provided, including an efficient procedure to optimize the hyperparameters in the model fitting, and paves the way for long-time large-scale simulations of such systems.
An electrostatic spectral neighbor analysis potential for lithium nitride
Machine-learned interatomic potentials based on local environment descriptors represent a transformative leap over traditional potentials based on rigid functional forms in terms of prediction
Machine learning based interatomic potential for amorphous carbon
We introduce a Gaussian approximation potential (GAP) for atomistic simulations of liquid and amorphous elemental carbon. Based on a machine learning representation of the density-functional theory
Machine Learning Predictions of Molecular Properties: Accurate Many-Body Potentials and Nonlocality in Chemical Space
A systematic hierarchy of efficient empirical methods to estimate atomization and total energies of molecules and is achieved by a vectorized representation of molecules (so-called Bag of Bonds model) that exhibits strong nonlocality in chemical space.
Development of a machine learning potential for graphene
© 2018 American Physical Society. We present an accurate interatomic potential for graphene, constructed using the Gaussian approximation potential (GAP) machine learning methodology. This GAP model
An Electrostatic Spectral Neighbor Analysis Potential (eSNAP) for Lithium Nitride
Machine-learned interatomic potentials based on local environment descriptors represent a transformative leap over traditional potentials based on rigid functional forms in terms of prediction