An orbital-based representation for accurate quantum machine learning
- Author(s)
- Konstantin Karandashev, O. Anatole Von Lilienfeld
- Abstract
We introduce an electronic structure based representation for quantum machine learning (QML) of electronic properties throughout chemical compound space. The representation is constructed using computationally inexpensive ab initio calculations and explicitly accounts for changes in the electronic structure. We demonstrate the accuracy and flexibility of resulting QML models when applied to property labels, such as total potential energy, HOMO and LUMO energies, ionization potential, and electron affinity, using as datasets for training and testing entries from the QM7b, QM7b-T, QM9, and LIBE libraries. For the latter, we also demonstrate the ability of this approach to account for molecular species of different charge and spin multiplicity, resulting in QML models that infer total potential energies based on geometry, charge, and spin as input.
- Organisation(s)
- Computational Materials Physics
- External organisation(s)
- Universität Basel
- Journal
- Journal of Chemical Physics
- Volume
- 156
- No. of pages
- 11
- ISSN
- 0021-9606
- DOI
- https://doi.org/10.1063/5.0083301
- Publication date
- 03-2022
- Peer reviewed
- Yes
- Austrian Fields of Science 2012
- 103006 Chemical physics
- Keywords
- ASJC Scopus subject areas
- General Physics and Astronomy, Physical and Theoretical Chemistry
- Portal url
- https://ucrisportal.univie.ac.at/en/publications/584fcf30-11dd-498a-a017-93dc69777fb0