Ksenia R. Briling^1*, Yannick Calvino Alonso^1*, Alberto Fabrizio^1*, Clemence Corminboeuf^1*

1 Institut des Sciences et Ingénierie Chimiques, École Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland

* Corresponding authors emails: ksenia.briling@epfl.ch, yannick.calvinoalonso@epfl.ch, alberto.fabrizio@epfl.ch, clemence.corminboeuf@epfl.ch

DOI10.24435/materialscloud:1g-w5 [version v1]

Publication date: Dec 08, 2023

How to cite this record

Ksenia R. Briling, Yannick Calvino Alonso, Alberto Fabrizio, Clemence Corminboeuf, SPAᴴM(a,b): encoding the density information from guess Hamiltonian in quantum machine learning representations, Materials Cloud Archive 2023.192 (2023), https://doi.org/10.24435/materialscloud:1g-w5

Description

Recently, we introduced a class of molecular representations for kernel-based regression methods — the spectrum of approximated Hamiltonian matrices (SPAᴴM) — that takes advantage of lightweight one-electron Hamiltonians traditionally used as an SCF initial guess. The original SPAᴴM variant is built from occupied-orbital energies (\ie, eigenvalues) and naturally contains all the information about nuclear charges, atomic positions, and symmetry requirements. Its advantages were demonstrated on datasets featuring a wide variation of charge and spin, for which traditional structure-based representations commonly fail. SPAᴴM(a,b), as introduced here, expands eigenvalue SPAᴴM into local and transferable representations. It relies upon one-electron density matrices to build fingerprints from atomic or bond density overlap contributions inspired from preceding state-of-the-art representations. The performance and efficiency of SPAᴴM(a,b) is assessed on the predictions for datasets of prototypical organic molecules (QM7) of different charges and azoheteroarene dyes in an excited state. Overall, both SPAᴴM(a) and SPAᴴM(b) outperform state-of-the-art representations on difficult prediction tasks such as the atomic properties of charged open-shell species and of π-conjugated systems.

Materials Cloud sections using this data

Files

File name	Size	Description
SPAHM+.tar.gz MD5md5:fc76968932247b7f4473696b45c6be86	1.6 GiB	Tar ball containing the geometries and the properties of all the datasets included in the manuscript, as well as the SPAHM(a,b) representation for neutral compounds in a binary format (for further details on the structure and the content of the tar ball, see README.txt)
README.txt MD5md5:5d0fc2baddea0c092274eb63e18204f9	3.0 KiB	README

License

Files and data are licensed under the terms of the following license: Creative Commons Attribution 4.0 International.
Metadata, except for email addresses, are licensed under the Creative Commons Attribution Share-Alike 4.0 International license.

External references

Keywords

machine learning SPAHM Guess Hamiltonians Molecular Representation EPFL MARVEL/P2 SNSF ERC