SPAᴴM(a,b): encoding the density information from guess Hamiltonian in quantum machine learning representations
JSON Export
{
"id": "2009",
"updated": "2024-01-10T08:51:53.596373+00:00",
"metadata": {
"version": 1,
"contributors": [
{
"givennames": "Ksenia",
"affiliations": [
"Institut des Sciences et Ing\u00e9nierie Chimiques, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland"
],
"email": "ksenia.briling@epfl.ch",
"familyname": "R. Briling"
},
{
"givennames": "Yannick",
"affiliations": [
"Institut des Sciences et Ing\u00e9nierie Chimiques, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland"
],
"email": "yannick.calvinoalonso@epfl.ch",
"familyname": "Calvino Alonso"
},
{
"givennames": "Alberto",
"affiliations": [
"Institut des Sciences et Ing\u00e9nierie Chimiques, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland"
],
"email": "alberto.fabrizio@epfl.ch",
"familyname": "Fabrizio"
},
{
"givennames": "Clemence",
"affiliations": [
"Institut des Sciences et Ing\u00e9nierie Chimiques, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland"
],
"email": "clemence.corminboeuf@epfl.ch",
"familyname": "Corminboeuf"
}
],
"title": "SPA\u1d34M(a,b): encoding the density information from guess Hamiltonian in quantum machine learning representations",
"_oai": {
"id": "oai:materialscloud.org:2009"
},
"keywords": [
"machine learning",
"SPAHM",
"Guess Hamiltonians",
"Molecular Representation",
"EPFL",
"MARVEL/P2",
"SNSF",
"ERC"
],
"publication_date": "Dec 08, 2023, 17:50:30",
"_files": [
{
"key": "SPAHM+.tar.gz",
"description": "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)",
"checksum": "md5:fc76968932247b7f4473696b45c6be86",
"size": 1708678712
},
{
"key": "README.txt",
"description": "README",
"checksum": "md5:5d0fc2baddea0c092274eb63e18204f9",
"size": 3024
}
],
"references": [
{
"doi": "10.1021/acs.jctc.3c01040",
"citation": "K. R. Briling, Y. Calvino Alonso, A. Fabrizio, and C. Corminboeuf, Preprint (2023)",
"url": "https://doi.org/10.1021/acs.jctc.3c01040",
"type": "Journal reference"
}
],
"description": "Recently, we introduced a class of molecular representations for kernel-based regression methods \u2014 the spectrum of approximated Hamiltonian matrices (SPA\u1d34M) \u2014 that takes advantage of lightweight one-electron Hamiltonians traditionally used as an SCF initial guess. The original SPA\u1d34M 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. SPA\u1d34M(a,b), as introduced here, expands eigenvalue SPA\u1d34M 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 SPA\u1d34M(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 SPA\u1d34M(a) and SPA\u1d34M(b) outperform state-of-the-art representations on difficult prediction tasks such as the atomic properties of charged open-shell species and of \u03c0-conjugated systems.",
"status": "published",
"license": "Creative Commons Attribution 4.0 International",
"conceptrecid": "2008",
"is_last": true,
"mcid": "2023.192",
"edited_by": 1210,
"id": "2009",
"owner": 1210,
"license_addendum": null,
"doi": "10.24435/materialscloud:1g-w5"
},
"revision": 5,
"created": "2023-12-04T23:48:46.547402+00:00"
}