Completing density functional theory by machine learning hidden messages from molecules
Дополнение теории функционала плотности с помощью машинного обучения скрытых сообщений от молекул
2020-05-05
SCID: 54.1/7ka2qu2m
Discuss with AI
Kohn–Sham density functional theoryexchange-correlation energy functionalfeed-forward neural networkmachine learning functionalnonlocal density descriptor
Figures from the paper
Abstract (AI)
Abstract Kohn–Sham density functional theory (DFT) is the basis of modern computational approaches to electronic structures. Their accuracy heavily relies on the exchange-correlation energy functional, which encapsulates electron–electron interaction beyond the classical model. As its universal form remains undiscovered, approximated functionals constructed with heuristic approaches are used for practical studies. However, there are problems in their accuracy and transferability, while any systematic approach to improve them is yet obscure. In this study, we demonstrate that the functional can be systematically constructed using accurate density distributions and energies in reference molecules via machine learning. Surprisingly, a trial functional machine learned from only a few molecules is already applicable to hundreds of molecules comprising various first- and second-row elements with the same accuracy as the standard functionals. This is achieved by relating density and energy using a flexible feed-forward neural network, which allows us to take a functional derivative via the back-propagation algorithm. In addition, simply by introducing a nonlocal density descriptor, the nonlocal effect is included to improve accuracy, which has hitherto been impractical. Our approach thus will help enrich the DFT framework by utilizing the rapidly advancing machine-learning technique.
Key Findings
1
A Kohn–Sham exchange-correlation functional can be systematically constructed by machine learning from accurate reference densities and energies.
2
A feed-forward neural network was used to relate electron density to energy and to obtain functional derivatives via back-propagation.
3
A trial functional trained on only a few molecules generalizes to hundreds of molecules containing various first- and second-row elements with accuracy comparable to standard functionals.
4
Introducing a nonlocal density descriptor into the learned functional incorporates nonlocal effects and improves accuracy previously impractical to include.
Research Object
Kohn–Sham density functional (exchange-correlation) as learned/constructed via machine learning from reference molecular densities and energies
Research Subject
Systematic construction and improvement of the exchange-correlation functional by relating electron density and energy with a neural network (including taking functional derivatives and incorporating a nonlocal density descriptor) to achieve transferable accuracy across many molecules
Publication Details
Publication Date
2020-05-05
Journal
Publisher
ISSN
Cited by
237
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai6
Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density1988
The M06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements: two new functionals and systematic testing of four M06-class functionals and 12 other functionals2007
Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields1994
A multicenter numerical integration scheme for polyatomic molecules1988
Efficient molecular numerical integration schemes1995
Density functional theory is straying from the path toward the exact functional2017