Atom-centered symmetry functions for constructing high-dimensional neural network potentials.

@article{Behler2011AtomcenteredSF,
  title={Atom-centered symmetry functions for constructing high-dimensional neural network potentials.},
  author={J{\"o}rg Behler},
  journal={The Journal of chemical physics},
  year={2011},
  volume={134 7},
  pages={
          074106
        }
}
  • J. Behler
  • Published 16 February 2011
  • Computer Science
  • The Journal of chemical physics
Neural networks offer an unbiased and numerically very accurate approach to represent high-dimensional ab initio potential-energy surfaces. Once constructed, neural network potentials can provide the energies and forces many orders of magnitude faster than electronic structure calculations, and thus enable molecular dynamics simulations of large systems. However, Cartesian coordinates are not a good choice to represent the atomic positions, and a transformation to symmetry functions is required… 
High-Dimensional Neural Network Potentials for Atomistic Simulations
High-dimensional neural network potentials, proposed by Behler and Parrinello in 2007, have become an established method to calculate potential energy surfaces with first-principles accuracy at a
Construction of high-dimensional neural network potentials using environment-dependent atom pairs.
TLDR
This work presents an implementation of an NN method based on atom pairs, and its accuracy and performance are compared to the atom-based NN approach, with the pair-based method yielding a slightly higher accuracy making it a competitive alternative for addressing complex systems in MD simulations.
Potential energy surfaces from high fidelity fitting of ab initio points: the permutation invariant polynomial - neural network approach
TLDR
This Review discusses the recent work on a new potential-fitting approach based on artificial neural networks, which are ultra-flexible in representing any multidimensional real functions, and its applications to the construction of multi-dimensional potential energy surfaces in many gas phase and gas–surface systems.
High-dimensional neural network potentials for magnetic systems using spin-dependent atom-centered symmetry functions
Machine learning potentials have emerged as a powerful tool to extend the time and length scales of first-principles quality simulations. Still, most machine learning potentials cannot distinguish
Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations.
  • J. Behler
  • Physics
    Physical chemistry chemical physics : PCCP
  • 2011
TLDR
In this Perspective, the current status of NN potentials is reviewed, and their advantages and limitations are discussed.
A neural network potential-energy surface for the water dimer based on environment-dependent atomic energies and charges.
TLDR
A full-dimensional NN potential for the water dimer is presented as a first step towards the construction of a NN Potential for liquid water, and it is shown that the potential and derived properties like vibrational frequencies are in excellent agreement with the underlying reference density-functional theory calculations.
Neural Network Potential Energy Surfaces for Small Molecules and Reactions.
TLDR
This work considers NN-based approaches to build PESs in the sums-of-product form important for quantum dynamics, ways to treat symmetry, and issues related to sampling data distributions and the relation between PES errors and errors in observables.
Library-Based LAMMPS Implementation of High-Dimensional Neural Network Potentials.
TLDR
A library of functions developed for the implementation of neural network potentials within the molecular dynamics package LAMMPS is presented and its performance using liquid water as a test system is demonstrated.
Representing potential energy surfaces by high-dimensional neural network potentials.
  • J. Behler
  • Materials Science
    Journal of physics. Condensed matter : an Institute of Physics journal
  • 2014
TLDR
The basic methodology of high-dimensional NNPs will be presented with a special focus on the scope and the remaining limitations of this approach, e.g. for addressing problems in materials science, for investigating properties of interfaces, and for studying solvation processes.
A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer
TLDR
This work introduces a fourth-generation high-dimensional neural network potential that combines a charge equilibration scheme employing environment-dependent atomic electronegativities with accurate atomic energies and substantially extends the applicability of modern machine learning potentials.
...
...

References

SHOWING 1-10 OF 48 REFERENCES
Generalized neural-network representation of high-dimensional potential-energy surfaces.
TLDR
A new kind of neural-network representation of DFT potential-energy surfaces is introduced, which provides the energy and forces as a function of all atomic positions in systems of arbitrary size and is several orders of magnitude faster than DFT.
Representing molecule-surface interactions with symmetry-adapted neural networks.
TLDR
This work builds the NN on a new type of symmetry functions, which allows to take the symmetry of the surface exactly into account and is illustrated by the application to a six-dimensional PES describing the interaction of oxygen molecules with the Al(111) surface.
Ab initio molecular dynamics of hydrogen dissociation on metal surfaces using neural networks and novelty sampling.
We outline a hybrid multiscale approach for the construction of ab initio potential energy surfaces (PESs) useful for performing six-dimensional (6D) classical or quantum mechanical molecular
A random-sampling high dimensional model representation neural network for building potential energy surfaces.
TLDR
It is shown that the number of available potential points determines the order of the HDMR which should be used and it is verified that it is possible to determine an accurate many-dimensional potential by doing low dimensional fits.
Using redundant coordinates to represent potential energy surfaces with lower-dimensional functions.
TLDR
A method for fitting potential energy surfaces with a sum of component functions of lower dimensionality that facilitates quantum dynamics calculations and reduces the dimensionality of the component functions by introducing new and redundant coordinates obtained with linear transformations from a neural network.
Neural network models of potential energy surfaces
Neural networks provide an efficient, general interpolation method for nonlinear functions of several variables. This paper describes the use of feed‐forward neural networks to model global
Representation of Intermolecular Potential Functions by Neural Networks
We have investigated how a neural network representation of intermolecular potential functions can be used to elevate some of the problems commonly encountered during fitting and application of
Ab initio quality neural-network potential for sodium
An interatomic potential for high-pressure high-temperature (HPHT) crystalline and liquid phases of sodium is created using a neural-network (NN) representation of the ab initio potential-energy
Descriptions of surface chemical reactions using a neural network representation of the potential-energy surface
A neural network NN approach is proposed for the representation of six-dimensional ab initio potentialenergy surfaces PES for the dissociation of a diatomic molecule at surfaces. We report tests of
...
...