We introduce a novel class of localized atomic environment representations based upon the Coulomb matrix. By combining these functions with the Gaussian approximation potential approach, we present LC-GAP, a new system for generating atomic potentials through machine learning (ML). Tests on the QM7, QM7b and GDB9 biomolecular datasets demonstrate that potentials created with LC-GAP can successfully predict atomization energies for molecules larger than those used for training to chemical accuracy, and can (in the case of QM7b) also be used to predict a range of other atomic properties with accuracy in line with the recent literature. As the best-performing representation has only linear dimensionality in the number of atoms in a local atomic environment, this represents an improvement in both prediction accuracy and computational cost when compared to similar Coulomb matrix-based methods.
CITATION STYLE
Barker, J., Bulin, J., Hamaekers, J., & Mathias, S. (2017). LC-GAP: Localized coulomb descriptors for the gaussian approximation potential. In Scientific Computing and Algorithms in Industrial Simulations: Projects and Products of Fraunhofer SCAI (pp. 25–42). Springer International Publishing. https://doi.org/10.1007/978-3-319-62458-7_2
Mendeley helps you to discover research relevant for your work.