Compressing local atomic neighbourhood descriptors.

4^th March 2022, 14:00
James Darby
University of Cambridge

Abstract

Many atomic descriptors are currently limited by their unfavourable scaling with the number of chemical elements S. For instance, the length of body-ordered descriptors typically scales as (NS)^v where v+1 is the body-order and N is the number of radial basis functions used in the density expansion. We introduce two distinct approaches which can be used to overcome this scaling for the Smooth Overlap of Atomic Positions (SOAP) power spectrum. Firstly, we show that the power spectrum is amenable to lossless compression with respect to both S and N, so that the descriptor length can be reduced from O(N^2 S^2) to O(NS). Secondly, we introduce a generalized SOAP kernel, where compression is achieved through the use of the total, element agnostic density, in combination with radial projection. The ideas used in the generalized kernel are equally applicably to any other body-ordered descriptors and we demonstrate this for the Atom Centered Symmetry Functions (ACSF). Finally, both compression approaches are shown to offer comparable performance to the original descriptor across a variety of numerical tests.