Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions

K. T. Schütt, M. Gastegger, A. Tkatchenko, K. R. Müller, R. J. Maurer

Research output: Contribution to journalArticle

Abstract

Machine learning advances chemistry and materials science by enabling large-scale exploration of chemical space based on quantum chemical calculations. While these models supply fast and accurate predictions of atomistic chemical properties, they do not explicitly capture the electronic degrees of freedom of a molecule, which limits their applicability for reactive chemistry and chemical analysis. Here we present a deep learning framework for the prediction of the quantum mechanical wavefunction in a local basis of atomic orbitals from which all other ground-state properties can be derived. This approach retains full access to the electronic structure via the wavefunction at force-field-like efficiency and captures quantum mechanics in an analytically differentiable representation. On several examples, we demonstrate that this opens promising avenues to perform inverse design of molecular structures for targeting electronic property optimisation and a clear path towards increased synergy of machine learning and quantum chemistry.

Original languageEnglish
Article number5024
JournalNature communications
Volume10
Issue number1
DOIs
Publication statusPublished - 2019 Dec 1

Fingerprint

Quantum chemistry
machine learning
quantum chemistry
Wave functions
Learning systems
chemistry
Quantum theory
Materials science
Electronic properties
Chemical properties
Ground state
Molecular structure
Electronic structure
Space Flight
materials science
predictions
Molecular Structure
Mechanics
chemical analysis
electronics

ASJC Scopus subject areas

  • Chemistry(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Physics and Astronomy(all)

Cite this

Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions. / Schütt, K. T.; Gastegger, M.; Tkatchenko, A.; Müller, K. R.; Maurer, R. J.

In: Nature communications, Vol. 10, No. 1, 5024, 01.12.2019.

Research output: Contribution to journalArticle

Schütt, K. T. ; Gastegger, M. ; Tkatchenko, A. ; Müller, K. R. ; Maurer, R. J. / Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions. In: Nature communications. 2019 ; Vol. 10, No. 1.
@article{144044b01c3243aa8f8936aad054e84c,
title = "Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions",
abstract = "Machine learning advances chemistry and materials science by enabling large-scale exploration of chemical space based on quantum chemical calculations. While these models supply fast and accurate predictions of atomistic chemical properties, they do not explicitly capture the electronic degrees of freedom of a molecule, which limits their applicability for reactive chemistry and chemical analysis. Here we present a deep learning framework for the prediction of the quantum mechanical wavefunction in a local basis of atomic orbitals from which all other ground-state properties can be derived. This approach retains full access to the electronic structure via the wavefunction at force-field-like efficiency and captures quantum mechanics in an analytically differentiable representation. On several examples, we demonstrate that this opens promising avenues to perform inverse design of molecular structures for targeting electronic property optimisation and a clear path towards increased synergy of machine learning and quantum chemistry.",
author = "Sch{\"u}tt, {K. T.} and M. Gastegger and A. Tkatchenko and M{\"u}ller, {K. R.} and Maurer, {R. J.}",
year = "2019",
month = "12",
day = "1",
doi = "10.1038/s41467-019-12875-2",
language = "English",
volume = "10",
journal = "Nature Communications",
issn = "2041-1723",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions

AU - Schütt, K. T.

AU - Gastegger, M.

AU - Tkatchenko, A.

AU - Müller, K. R.

AU - Maurer, R. J.

PY - 2019/12/1

Y1 - 2019/12/1

N2 - Machine learning advances chemistry and materials science by enabling large-scale exploration of chemical space based on quantum chemical calculations. While these models supply fast and accurate predictions of atomistic chemical properties, they do not explicitly capture the electronic degrees of freedom of a molecule, which limits their applicability for reactive chemistry and chemical analysis. Here we present a deep learning framework for the prediction of the quantum mechanical wavefunction in a local basis of atomic orbitals from which all other ground-state properties can be derived. This approach retains full access to the electronic structure via the wavefunction at force-field-like efficiency and captures quantum mechanics in an analytically differentiable representation. On several examples, we demonstrate that this opens promising avenues to perform inverse design of molecular structures for targeting electronic property optimisation and a clear path towards increased synergy of machine learning and quantum chemistry.

AB - Machine learning advances chemistry and materials science by enabling large-scale exploration of chemical space based on quantum chemical calculations. While these models supply fast and accurate predictions of atomistic chemical properties, they do not explicitly capture the electronic degrees of freedom of a molecule, which limits their applicability for reactive chemistry and chemical analysis. Here we present a deep learning framework for the prediction of the quantum mechanical wavefunction in a local basis of atomic orbitals from which all other ground-state properties can be derived. This approach retains full access to the electronic structure via the wavefunction at force-field-like efficiency and captures quantum mechanics in an analytically differentiable representation. On several examples, we demonstrate that this opens promising avenues to perform inverse design of molecular structures for targeting electronic property optimisation and a clear path towards increased synergy of machine learning and quantum chemistry.

UR - http://www.scopus.com/inward/record.url?scp=85075114576&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075114576&partnerID=8YFLogxK

U2 - 10.1038/s41467-019-12875-2

DO - 10.1038/s41467-019-12875-2

M3 - Article

C2 - 31729373

AN - SCOPUS:85075114576

VL - 10

JO - Nature Communications

JF - Nature Communications

SN - 2041-1723

IS - 1

M1 - 5024

ER -