Machine learning for potential energy surfaces: An extensive database and assessment of methods

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

Standard

Machine learning for potential energy surfaces : An extensive database and assessment of methods. / Schmitz, Gunnar; Godtliebsen, Ian Heide; Christiansen, Ove.

I: The Journal of Chemical Physics, Bind 150, Nr. 24, 244113, 2019.

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisTidsskriftartikelForskningpeer review

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{98756e482286498e847494b0a7eb85e8,
title = "Machine learning for potential energy surfaces: An extensive database and assessment of methods",
abstract = "On the basis of a new extensive database constructed for the purpose, we assess various Machine Learning (ML) algorithms to predict energies in the framework of potential energy surface (PES) construction and discuss black box character, robustness, and efficiency. The database for training ML algorithms in energy predictions based on the molecular structure contains SCF, RI-MP2, RI-MP2-F12, and CCSD(F12 ∗)(T) data for around 10.5 × 10 6 configurations of 15 small molecules. The electronic energies as function of molecular structure are computed from both static and iteratively refined grids in the context of automized PES construction for anharmonic vibrational computations within the n-mode expansion. We explore the performance of a range of algorithms including Gaussian Process Regression (GPR), Kernel Ridge Regression, Support Vector Regression, and Neural Networks (NNs). We also explore methods related to GPR such as sparse Gaussian Process Regression, Gaussian process Markov Chains, and Sparse Gaussian Process Markov Chains. For NNs, we report some explorations of architecture, activation functions, and numerical settings. Different delta-learning strategies are considered, and the use of delta learning targeting CCSD(F12 ∗)(T) predictions using, for example, RI-MP2 combined with machine learned CCSD(F12 ∗)(T)-RI-MP2 differences is found to be an attractive option.",
keywords = "CHEMISTRY, CLUSTERS, FORCE-FIELDS, IDENTITY, MODEL, OPTIMIZATION, PREDICTION, REGRESSION, RESOLUTION, SELF-CONSISTENT-FIELD",
author = "Gunnar Schmitz and Godtliebsen, {Ian Heide} and Ove Christiansen",
year = "2019",
doi = "10.1063/1.5100141",
language = "English",
volume = "150",
journal = "Journal of Chemical Physics",
issn = "0021-9606",
publisher = "AMER INST PHYSICS",
number = "24",

}

RIS

TY - JOUR

T1 - Machine learning for potential energy surfaces

T2 - An extensive database and assessment of methods

AU - Schmitz, Gunnar

AU - Godtliebsen, Ian Heide

AU - Christiansen, Ove

PY - 2019

Y1 - 2019

N2 - On the basis of a new extensive database constructed for the purpose, we assess various Machine Learning (ML) algorithms to predict energies in the framework of potential energy surface (PES) construction and discuss black box character, robustness, and efficiency. The database for training ML algorithms in energy predictions based on the molecular structure contains SCF, RI-MP2, RI-MP2-F12, and CCSD(F12 ∗)(T) data for around 10.5 × 10 6 configurations of 15 small molecules. The electronic energies as function of molecular structure are computed from both static and iteratively refined grids in the context of automized PES construction for anharmonic vibrational computations within the n-mode expansion. We explore the performance of a range of algorithms including Gaussian Process Regression (GPR), Kernel Ridge Regression, Support Vector Regression, and Neural Networks (NNs). We also explore methods related to GPR such as sparse Gaussian Process Regression, Gaussian process Markov Chains, and Sparse Gaussian Process Markov Chains. For NNs, we report some explorations of architecture, activation functions, and numerical settings. Different delta-learning strategies are considered, and the use of delta learning targeting CCSD(F12 ∗)(T) predictions using, for example, RI-MP2 combined with machine learned CCSD(F12 ∗)(T)-RI-MP2 differences is found to be an attractive option.

AB - On the basis of a new extensive database constructed for the purpose, we assess various Machine Learning (ML) algorithms to predict energies in the framework of potential energy surface (PES) construction and discuss black box character, robustness, and efficiency. The database for training ML algorithms in energy predictions based on the molecular structure contains SCF, RI-MP2, RI-MP2-F12, and CCSD(F12 ∗)(T) data for around 10.5 × 10 6 configurations of 15 small molecules. The electronic energies as function of molecular structure are computed from both static and iteratively refined grids in the context of automized PES construction for anharmonic vibrational computations within the n-mode expansion. We explore the performance of a range of algorithms including Gaussian Process Regression (GPR), Kernel Ridge Regression, Support Vector Regression, and Neural Networks (NNs). We also explore methods related to GPR such as sparse Gaussian Process Regression, Gaussian process Markov Chains, and Sparse Gaussian Process Markov Chains. For NNs, we report some explorations of architecture, activation functions, and numerical settings. Different delta-learning strategies are considered, and the use of delta learning targeting CCSD(F12 ∗)(T) predictions using, for example, RI-MP2 combined with machine learned CCSD(F12 ∗)(T)-RI-MP2 differences is found to be an attractive option.

KW - CHEMISTRY

KW - CLUSTERS

KW - FORCE-FIELDS

KW - IDENTITY

KW - MODEL

KW - OPTIMIZATION

KW - PREDICTION

KW - REGRESSION

KW - RESOLUTION

KW - SELF-CONSISTENT-FIELD

UR - http://www.scopus.com/inward/record.url?scp=85068214401&partnerID=8YFLogxK

U2 - 10.1063/1.5100141

DO - 10.1063/1.5100141

M3 - Journal article

VL - 150

JO - Journal of Chemical Physics

JF - Journal of Chemical Physics

SN - 0021-9606

IS - 24

M1 - 244113

ER -