Description
Representing atomic neighbourhood environments play an important role in high-throughput materials modelling applications. For example, machine-learning based fitting of potential energy surfaces of atomic systems requires faithful representation of chemical environments. Such representations need to be invariant to rotations and permutations of identical atoms while changing in a continuous and smooth manner with the atomic positions. The author presents a unifying view of different approaches and examine their behaviour in concrete numerical examples. Finally, the author will introduce a new way of measuring similarity of atomic environments, which intends to eliminate the shortcomings of earlier representations.
Recommended Citation
Bartók-Pártay, A., & Csányi, G. (2014). Representing atomic environments. In A. Bajaj, P. Zavattieri, M. Koslowski, & T. Siegmund (Eds.). Proceedings of the Society of Engineering Science 51st Annual Technical Meeting, October 1-3, 2014 , West Lafayette: Purdue University Libraries Scholarly Publishing Services, 2014. https://docs.lib.purdue.edu/ses2014/mms/imms/6
Representing atomic environments
Representing atomic neighbourhood environments play an important role in high-throughput materials modelling applications. For example, machine-learning based fitting of potential energy surfaces of atomic systems requires faithful representation of chemical environments. Such representations need to be invariant to rotations and permutations of identical atoms while changing in a continuous and smooth manner with the atomic positions. The author presents a unifying view of different approaches and examine their behaviour in concrete numerical examples. Finally, the author will introduce a new way of measuring similarity of atomic environments, which intends to eliminate the shortcomings of earlier representations.