https://doi.org/10.1140/epjs/s11734-021-00095-z
Regular Article
Learning DFT
HQS Quantum Simulations GmbH, Haid-und-Neu-Straße 7, 76131, Karlsruhe, Germany
a
Peter.schmitteckert@quantumsimulations.de
Received:
21
July
2020
Accepted:
5
January
2021
Published online:
19
April
2021
We present an extension of reverse engineered Kohn-Sham potentials from a density matrix renormalization group calculation towards the construction of a density functional theory functional via deep learning. Instead of applying machine learning to the energy functional itself, we apply these techniques to the Kohn-Sham potentials. To this end, we develop a scheme to train a neural network to represent the mapping from local densities to Kohn-Sham potentials. Finally, we use the neural network to up-scale the simulation to larger system sizes.
© The Author(s), under exclusive licence to EDP Sciences, Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2021
