Incompleteness of Atomic Structure Representations.
- Pozdnyakov SN Laboratory of Computational Science and Modelling, Institute of Materials, Ecole Polytechnique Fédérale de Lausanne, Lausanne 1015, Switzerland.
- Willatt MJ Laboratory of Computational Science and Modelling, Institute of Materials, Ecole Polytechnique Fédérale de Lausanne, Lausanne 1015, Switzerland.
- Bartók AP Department of Physics and Warwick Centre for Predictive Modelling, School of Engineering, University of Warwick, Coventry CV4 7AL, United Kingdom.
- Ortner C Mathematics Institute, University of Warwick, Coventry CV4 7AL, United Kingdom.
- Csányi G Engineering Laboratory, University of Cambridge, Trumpington Street, Cambridge CB2 1PZ, United Kingdom.
- Ceriotti M Laboratory of Computational Science and Modelling, Institute of Materials, Ecole Polytechnique Fédérale de Lausanne, Lausanne 1015, Switzerland.
- 2020-10-30
Published in:
- Physical review letters. - 2020
English
Many-body descriptors are widely used to represent atomic environments in the construction of machine-learned interatomic potentials and more broadly for fitting, classification, and embedding tasks on atomic structures. There is a widespread belief in the community that three-body correlations are likely to provide an overcomplete description of the environment of an atom. We produce several counterexamples to this belief, with the consequence that any classifier, regression, or embedding model for atom-centered properties that uses three- (or four)-body features will incorrectly give identical results for different configurations. Writing global properties (such as total energies) as a sum of many atom-centered contributions mitigates the impact of this fundamental deficiency-explaining the success of current "machine-learning" force fields. We anticipate the issues that will arise as the desired accuracy increases, and suggest potential solutions.
- Language
-
- English
- Open access status
- green
- Identifiers
-
- DOI 10.1103/PhysRevLett.125.166001
- PMID 33124874
- Persistent URL
- https://sonar.ch/global/documents/180683
Statistics
Document views: 52
File downloads:
- Full-text: 0