Deep learning has led to a paradigm shift in artificial intelligence, including web, text, and image search, speech recognition, as well as bioinformatics, with growing impact in chemical physics. Machine learning, in general, and deep learning, in particular, are ideally suitable for representing quantum-mechanical interactions, enabling us to model nonlinear potential-energy surfaces or enhancing the exploration of chemical compound space. Here we present the deep learning architecture SchNet that is specifically designed to model atomistic systems by making use of continuous-filter convolutional layers. We demonstrate the capabilities of SchNet by accurately predicting a range of properties across chemical space for molecules and materials, where our model learns chemically plausible embeddings of atom types across the periodic table. Finally, we employ SchNet to predict potential-energy surfaces and energy-conserving force fields for molecular dynamics simulations of small molecules and perform an exemplary study on the quantum-mechanical properties of C20-fullerene that would have been infeasible with regular ab initio molecular dynamics.
Skip Nav Destination
Article navigation
28 June 2018
Research Article|
March 29 2018
SchNet – A deep learning architecture for molecules and materials
Special Collection:
Data-Enabled Theoretical Chemistry
K. T. Schütt
;
K. T. Schütt
a)
1
Machine Learning Group, Technische Universität Berlin
, 10587 Berlin, Germany
Search for other works by this author on:
H. E. Sauceda;
H. E. Sauceda
2
Fritz-Haber-Institut der Max-Planck-Gesellschaft
, 14195 Berlin, Germany
Search for other works by this author on:
P.-J. Kindermans;
P.-J. Kindermans
1
Machine Learning Group, Technische Universität Berlin
, 10587 Berlin, Germany
Search for other works by this author on:
A. Tkatchenko;
A. Tkatchenko
b)
3
Physics and Materials Science Research Unit, University of Luxembourg
, L-1511 Luxembourg, Luxembourg
Search for other works by this author on:
K.-R. Müller
K.-R. Müller
c)
1
Machine Learning Group, Technische Universität Berlin
, 10587 Berlin, Germany
4
Max-Planck-Institut für Informatik
, Saarbrücken, Germany
5
Department of Brain and Cognitive Engineering, Korea University
, Anam-dong, Seongbuk-gu, Seoul 136-713, South Korea
Search for other works by this author on:
J. Chem. Phys. 148, 241722 (2018)
Article history
Received:
December 16 2017
Accepted:
March 08 2018
Citation
K. T. Schütt, H. E. Sauceda, P.-J. Kindermans, A. Tkatchenko, K.-R. Müller; SchNet – A deep learning architecture for molecules and materials. J. Chem. Phys. 28 June 2018; 148 (24): 241722. https://doi.org/10.1063/1.5019779
Download citation file:
Pay-Per-View Access
$40.00
Sign In
You could not be signed in. Please check your credentials and make sure you have an active account and try again.
Citing articles via
DeePMD-kit v2: A software package for deep potential models
Jinzhe Zeng, Duo Zhang, et al.
CREST—A program for the exploration of low-energy molecular chemical space
Philipp Pracht, Stefan Grimme, et al.
Dielectric profile at the Pt(111)/water interface
Jia-Xin Zhu, Jun Cheng, et al.
Related Content
Reproducing global potential energy surfaces with continuous-filter convolutional neural networks
J. Chem. Phys. (May 2019)
Challenges for machine learning force fields in reproducing potential energy surfaces of flexible molecules
J. Chem. Phys. (March 2021)
A deep neural network for molecular wave functions in quasi-atomic minimal basis representation
J. Chem. Phys. (July 2020)