Atomic energy in grain boundaries studied by machine learning

Xin-yuan, Song; Deng, Chuang

doi:10.1103/physrevmaterials.6.043601

Cited by 11 publications

(6 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…42 ML models for other properties of local atomic environments have recently been investigated, ranging from the local electronic density of states 43,44 to local distortion factors in grain boundaries. 45 The nature of these local ML properties (including atomic energies), and their usefulness in predicting physical properties, remains an interesting research question. (See, e.g., Ref.…”

Section: Introductionmentioning

confidence: 99%

Exploring the configurational space of amorphous graphene with machine-learned atomic energies

2022

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

Exploring the configurational space of amorphous graphene with machine-learned atomic energies

2022

View full text Add to dashboard Cite

show abstract

“…Again, the overhead of performing a decomposition of total properties into their atomic constituents will depend on the exact scheme employed, but the cost may generally be assumed to be of the order of a single mean-field iteration. Such a strategy has previously been considered for solid-state applications, e.g., in the design of force fields , and in the study of grain boundaries . In here, we will instead restrict our focus to chemical Hamiltonians only, seeking to evaluate the usefulness of training standard NNs on atomic, rather than solely total energies.…”

Section: Introductionmentioning

confidence: 99%

“…Such a strategy has previously been considered for solid-state applications, e.g., in the design of force fields 20,21 and in the study of grain boundaries. 22 In here, we will instead restrict our focus to chemical Hamiltonians only, seeking to evaluate the usefulness of training standard NNs on atomic, rather than solely total energies. For this purpose, we will decompose the standard QM7 data set, 23,24 which consists of 7165 constitutional and structural molecular isomers each containing up to 7 heavy atoms (C, N, O, and S), with the largest molecule built from a mere total of 23 atoms.…”

Section: Introductionmentioning

confidence: 99%

Decomposing Chemical Space: Applications to the Machine Learning of Atomic Energies

Kjeldal

Eriksen

2023

J. Chem. Theory Comput.

View full text Add to dashboard Cite

We apply a number of atomic decomposition schemes across the standard QM7 data seta small model set of organic molecules at equilibrium geometryto inspect the possible emergence of trends among contributions to atomization energies from distinct elements embedded within molecules. Specifically, a recent decomposition scheme of ours based on spatially localized molecular orbitals is compared to alternatives that instead partition molecular energies on account of which nuclei individual atomic orbitals are centered on. We find these partitioning schemes to expose the composition of chemical compound space in very dissimilar ways in terms of the grouping, binning, and heterogeneity of discrete atomic contributions, e.g., those associated with hydrogens bonded to different heavy atoms. Furthermore, unphysical dependencies on the one-electron basis set are found for some, but not all of these schemes. The relevance and importance of these compositional factors for training tailored neural network models based on atomic energies are next assessed. We identify both limitations and possible advantages with respect to contemporary machine learning models and discuss the design of potential counterparts based on atoms and the intrinsic energies of these as the principal decomposition units.

show abstract

“…The ML method has been applied in many research fields [ 16 , 17 , 18 , 19 , 20 ], and provides an efficient technique by which to link the structure-property of a GB, particularly to extract correlations from high-dimensional datasets [ 21 , 22 , 23 , 24 , 25 ], and has been successfully applied in predicting GB energies [ 21 , 23 , 24 , 25 , 26 , 27 ], point defect segregation energies [ 28 , 29 ], GB structures [ 30 ] and damages and deformations in GB [ 31 , 32 ]. Usually, an appropriate ML method is employed according to the datasets and the expected correlations.…”

Section: Introductionmentioning

confidence: 99%

Standard Deviation Effect of Average Structure Descriptor on Grain Boundary Energy Prediction

Dang

2023

Materials

View full text Add to dashboard Cite

The structural complexities of grain boundaries (GBs) result in their complicated property contributions to polycrystalline metals and alloys. In this study, we propose a GB structure descriptor by linearly combining the average two-point correlation function (PCF) and standard deviation of PCF via a weight parameter, to reveal the standard deviation effect of PCF on energy predictions of Cu, Al and Ni asymmetric tilt GBs (i.e., Σ3, Σ5, Σ9, Σ11, Σ13 and Σ17), using two machine learning (ML) methods; i.e., principal component analysis (PCA)-based linear regression and recurrent neural networks (RNN). It is found that the proposed structure descriptor is capable of improving GB energy prediction for both ML methods. This suggests the discriminatory power of average PCF for different GBs is lifted since the proposed descriptor contains the data dispersion information. Meanwhile, we also show that GB atom selection methods by which PCF is evaluated also affect predictions.

show abstract

Atomic energy in grain boundaries studied by machine learning

Cited by 11 publications

References 45 publications

Exploring the configurational space of amorphous graphene with machine-learned atomic energies

Exploring the configurational space of amorphous graphene with machine-learned atomic energies

Decomposing Chemical Space: Applications to the Machine Learning of Atomic Energies

Standard Deviation Effect of Average Structure Descriptor on Grain Boundary Energy Prediction

Contact Info

Product

Resources

About