Abstract
Machine learning has been successfully applied to the prediction of chemical properties of small organic molecules such as energies or polarizabilities. Compared to these properties, the electronic excitation energies pose a much more challenging learning problem. Here, we examine the applicability of two existing machine learning methodologies to the prediction of excitation energies from time-dependent density functional theory. To this end, we systematically study the performance of various 2- and 3-body descriptors as well as the deep neural network SchNet to predict extensive as well as intensive properties such as the transition energies from the ground state to the first and second excited state. As perhaps expected current state-of-the-art machine learning techniques are more suited to predict extensive as opposed to intensive quantities. We speculate on the need to develop global descriptors that can describe both extensive and intensive properties on equal footing.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
M. Grätzel, Nature 414, 338 (2001)
M. Gross et al., Nature 405, 661 (2000)
E. Runge, E.K.U. Gross, Phys. Rev. Lett. 52, 997 (1984)
M. Rupp et al., Phys. Rev. Lett. 108, 058301 (2012)
K. Hansen et al., J. Chem. Theory Comput. 9, 3404 (2013)
G. Montavon et al., New J. Phys. 15, 095003 (2013)
F.A. Faber et al., J. Chem. Theory Comput. 13, 5255 (2017)
K.T. Schütt et al., Nat. Commun. 8, 13890 (2017)
K.T. Schütt et al., Adv. Neural Inf. Process. Syst. 30, 992 (2017)
K.T. Schütt et al., J. Chem. Phys. 148, 241722 (2018)
S. Chmiela et al., Sci. Adv. 3, e1603015 (2017)
F. Noé, C. Clementi, J. Chem. Theory Comput. 11, 5002 (2015)
M. Gastegger, J. Behler, P. Marquetand, Chem. Sci. 8, 6924 (2017)
A.P. Bartók et al., Phys. Rev. B 88, 054104 (2013)
A. Mardt et al., Nat. Commun. 9, 5 (2018)
J.C. Snyder et al., Phys. Rev. Lett. 108, 253002 (2012)
F. Brockherde et al., Nat. Commun. 8, 872 (2017)
C.R. Collins et al., arXiv:1701.06649 [physics.chem-ph] (2017)
B. Huang, O. Anatole von Lilienfeld, J. Chem. Phys. 145, 161102 (2016)
G. Montavon et al., in Advances in Neural Information Processing Systems, edited by F. Pereira et al. (Curran Associates, Inc., NewYork, 2012), Vol. 25, pp. 440–448
H. Huo, M. Rupp, arXiv:1704.06439 (2017)
J. Behler, M. Parrinello, Phys. Rev. Lett. 98, 146401 (2007)
A.P. Bartók et al., Sci. Adv. 3, e1701816 (2017)
A.P. Bartók et al., Phys. Rev. Lett. 104, 136403 (2010)
K. Hansen et al., J. Phys. Chem. Lett. 6, 2326 (2015)
W. Pronobis, A. Tkatchenko, K.-R. Müller, J. Chem. Theory Comput. 14, 2991 (2018)
R. Ramakrishnan et al., Sci. Data 1, 140022 EP (2014)
R. Ramakrishnan et al., J. Chem. Phys. 143, 084111 (2015)
F. Furche, R. Ahlrichs, J. Chem. Phys. 117, 7433 (2002)
J.P. Perdew, M. Ernzerhof, K. Burke, J. Chem. Phys. 105, 9982 (1996)
C. Adamo, V. Barone, J. Chem. Phys. 110, 6158 (1999)
F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7, 3297 (2005)
C.M. Bishop, Pattern Recognition and Machine Learning (Information Science and Statistics) (Springer-Verlag, Inc., Secaucus, NJ, USA, New York, 2006)
K.-R. Müller et al., IEEE Trans. Neural Netw. 12, 181 (2001)
Author information
Authors and Affiliations
Corresponding authors
Additional information
Contribution to the Topical Issue “Special issue in honor of Hardy Gross”, edited by C.A. Ullrich, F.M.S. Nogueira, A. Rubio, and M.A.L. Marques.
Rights and permissions
Open Access This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
About this article
Cite this article
Pronobis, W., Schütt, K.T., Tkatchenko, A. et al. Capturing intensive and extensive DFT/TDDFT molecular properties with machine learning. Eur. Phys. J. B 91, 178 (2018). https://doi.org/10.1140/epjb/e2018-90148-y
Received:
Revised:
Published:
DOI: https://doi.org/10.1140/epjb/e2018-90148-y