Manifolds of quasiconstant SOAP and ACSF fingerprints and the resulting failure to machine learn fourbody interactions
Abstract
Atomic fingerprints are commonly used for the characterization of local environments of atoms in machine learning and other contexts. In this work, we study the behavior of two widely used fingerprints, namely, the smooth overlap of atomic positions (SOAP) and the atomcentered symmetry functions (ACSFs), under finite changes of atomic positions and demonstrate the existence of manifolds of quasiconstant fingerprints. These manifolds are found numerically by following eigenvectors of the sensitivity matrix with quasizero eigenvalues. The existence of such manifolds in ACSF and SOAP causes a failure to machine learn fourbody interactions, such as torsional energies that are part of standard force fields. No such manifolds can be found for the overlap matrix (OM) fingerprint due to its intrinsic manybody character.
 Publication:

Journal of Chemical Physics
 Pub Date:
 January 2022
 DOI:
 10.1063/5.0070488
 arXiv:
 arXiv:2102.06915
 Bibcode:
 2022JChPh.156c4302P
 Keywords:

 Condensed Matter  Other Condensed Matter;
 Physics  Computational Physics
 EPrint:
 8 pages, 5 figures