DFT study of NO and H2O co-adsorption on Cu Co (m+n=2∼7) clusters

Huo, Pei-Ying; Zhang, Xiurong; Yu, Zhicheng; Gao, Kun

doi:10.1016/j.molstruc.2017.07.062

Cited by 6 publications

(5 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Symmetry aside, progress can be made by attempting to use kernels based on simple, descriptive features corresponding to low-dimensional feature spaces. Taking inspiration from parametrized force fields, these descriptors could e.g., be chosen to be interatomic distances taken singularly or in triplets, yielding kernels based on 2-or 3-body interactions [16,31,32]. Since low-dimensional feature spaces allow efficient learning (convergence is reached using small databases), to the extent that simple descriptors capture the correct physics, the GP process will be a relatively fast, while still very accurate, interpolator.…”

Section: B Prior Knowledge and Gp Kernelsmentioning

confidence: 99%

See 1 more Smart Citation

Efficient nonparametricn-body force fields from machine learning

2018

View full text Add to dashboard Cite

We provide a definition and explicit expressions for n-body Gaussian Process (GP) kernels, which can learn any interatomic interaction occurring in a physical system, up to n-body contributions, for any value of n. The series is complete, as it can be shown that the "universal approximator" squared exponential kernel can be written as a sum of n-body kernels. These recipes enable the choice of optimally efficient force models for each target system, as confirmed by extensive testing on various materials. We furthermore describe how the n-body kernels can be "mapped" on equivalent representations that provide database-size-independent predictions and are thus crucially more efficient. We explicitly carry out this mapping procedure for the first non-trivial (3-body) kernel of the series, and we show that this reproduces the GP-predicted forces with meV/Å accuracy while being orders of magnitude faster. These results pave the way to using novel force models (here named "M-FFs") that are computationally as fast as their corresponding standard parametrised n-body force fields, while retaining the nonparametric character, the ease of training and validation, and the accuracy of the best recently proposed machine learning potentials.

show abstract

Section: B Prior Knowledge and Gp Kernelsmentioning

confidence: 99%

“…This kernel has a full many-body character [28], ensured by the prescribed normalisation step defined by Eqs. (31)(32)(33)(34)(35)(36) of the standard Ref. [28]), which made it possible to use it e.g., to augment to full many-body the descriptive power of a 2and 3-body explicit kernel expansion [26].…”

Section: Building N-body Kernels I: So(3) Integrationmentioning

confidence: 99%

Efficient nonparametricn-body force fields from machine learning

2018

View full text Add to dashboard Cite

show abstract

“…Several atomic structure descriptors for machine-learning models have been proposed in the literature. [53][54][55][56][57][58][59] Here we employ a recently developed descriptor based on the expansion of the radial and angular distribution functions (RDF and ADF) that is numerically efficient and has the advantage that its complexity does not increase with the number of chemical species. [60] A complete dervation and benchmarks of the method can be found in the original reference 60.…”

Section: Descriptor Of the Local Atomic Environmentmentioning

confidence: 99%

Constructing first-principles phase diagrams of amorphous LixSi using machine-learning-assisted sampling with an evolutionary algorithm

Artrith

Urban

Ceder

2018

The Journal of Chemical Physics

159

132

View full text Add to dashboard Cite

The atomistic modeling of amorphous materials requires structure sizes and sampling statistics that are challenging to achieve with first-principles methods. Here, we propose a methodology to speed up the sampling of amorphous and disordered materials using a combination of a genetic algorithm and a specialized machine-learning potential based on artificial neural networks (ANNs). We show for the example of the amorphous LiSi alloy that around 1000 first-principles calculations are sufficient for the ANN-potential assisted sampling of low-energy atomic configurations in the entire amorphous LiSi phase space. The obtained phase diagram is validated by comparison with the results from an extensive sampling of LiSi configurations using molecular dynamics simulations and a general ANN potential trained to ∼45 000 first-principles calculations. This demonstrates the utility of the approach for the first-principles modeling of amorphous materials.

show abstract

“…This scaling has so far limited current MLP approaches to compositions with only a few chemical species [7][8][9][10] or atomic structures [11]. Overcoming this limitation is a very active field of research [12,13].In this article we demonstrate that the computational complexity of MLPs does not necessarily grow with the number of chemical species, so that MLPs for materials with ten or more chemical species are in principle feasible and computationally efficient. We show that, contrary to intuition and common belief, the same model complexity that is optimal for a ternary material is also sufficient to describe a system with 11 chemical species (Fig.…”

mentioning

confidence: 94%

mentioning

confidence: 99%

Efficient and accurate machine-learning interpolation of atomic energies in compositions with many species

2017

View full text Add to dashboard Cite

Machine-learning potentials (MLPs) for atomistic simulations are a promising alternative to conventional classical potentials. Current approaches rely on descriptors of the local atomic environment with dimensions that increase quadratically with the number of chemical species. In this article, we demonstrate that such a scaling can be avoided in practice. We show that a mathematically simple and computationally efficient descriptor with constant complexity is sufficient to represent transition-metal oxide compositions and biomolecules containing 11 chemical species with a precision of around 3 meV/atom. This insight removes a perceived bound on the utility of MLPs and paves the way to investigate the physics of previously inaccessible materials with more than ten chemical species.

show abstract

DFT study of NO and H2O co-adsorption on Cu Co (m+n=2∼7) clusters

Cited by 6 publications

References 34 publications

Efficient nonparametricn-body force fields from machine learning

Efficient nonparametricn-body force fields from machine learning

Constructing first-principles phase diagrams of amorphous LixSi using machine-learning-assisted sampling with an evolutionary algorithm

Efficient and accurate machine-learning interpolation of atomic energies in compositions with many species

Contact Info

Product

Resources

About