Solid harmonic wavelet scattering for predictions of molecule properties

Michael Eickenberg, Georgios Exarchakis, Matthew Hirn, Stéphane Mallat, Louis Thiry

Research output: Contribution to journalArticlepeer-review

53 Citations (SciVal)

Abstract

We present a machine learning algorithm for the prediction of molecule properties inspired by ideas from density functional theory (DFT). Using Gaussian-type orbital functions, we create surrogate electronic densities of the molecule from which we compute invariant "solid harmonic scattering coefficients" that account for different types of interactions at different scales. Multilinear regressions of various physical properties of molecules are computed from these invariant coefficients. Numerical experiments show that these regressions have near state-of-the-art performance, even with relatively few training examples. Predictions over small sets of scattering coefficients can reach a DFT precision while being interpretable.

Original languageEnglish
Article number241732
JournalJournal of Chemical Physics
Volume148
Issue number24
DOIs
Publication statusPublished - 28 Jun 2018

Bibliographical note

Funding Information:
M.E., G.E., and S.M. are supported by ERC Grant Invari-antClass No. 320959; M.H. is supported by the Alfred P. Sloan Fellowship (Grant No. FG-2016-6607), the DARPA YFA (Grant No. D16AP00117), and NSF Grant No. 1620216.

Funding Information:
M.E., G.E., and S.M. are supported by ERC Grant InvariantClass No. 320959; M.H. is supported by the Alfred P. Sloan Fellowship (Grant No. FG-2016-6607), the DARPA YFA (Grant No. D16AP00117), and NSF Grant No. 1620216.

Publisher Copyright:
© 2018 Author(s).

ASJC Scopus subject areas

  • General Physics and Astronomy
  • Physical and Theoretical Chemistry

Fingerprint

Dive into the research topics of 'Solid harmonic wavelet scattering for predictions of molecule properties'. Together they form a unique fingerprint.

Cite this