PreprintCC BY 4.0arXiv
ROSE: Deep Learning exchange-correlation functional from reduced density-matrix representations
Abstract
We introduce ROSE, a deep-learning exchange--correlation energy functional built from the occupied-state density matrix. On GMTKN55 and on additional external benchmarks considered in this work, ROSE improves over modern neural network functionals trained under matched protocols and is on average more accurate than, or on par with, most hybrid and range-separated hybrid functionals considered here. On the experimental benchmarks considered here it has leading accuracy across hybrids, range separated hybrids, and neural functionals. These results motivate nonlocal density-matrix representations as a route to systematic improvement of learned XC functionals with low computational cost.
Related
- Machine-learned Laplacian-level density functional from exact exchange-correlation potentials and energies
- Toward resolving the RuO₂/IrO₂ oxygen evolution activity puzzle: critical roles of spin polarization and exchange-correlation functional accuracy.
- Molecular Implementation of the Machine-Learned Skala Exchange-Correlation Functional in CP2K through GauXC
- Native implementation of the machine-learned Skala exchange-correlation functional in CP2K: Unified one-centre reconstruction for molecular and condensed-phase calculations
- Nonempirical Time-Dependent Density Functional Theory Framework for Nonlocal Exchange--Correlation Potentials
BibTeX
@misc{kulaev2026rose,
title = {{ROSE: Deep Learning exchange-correlation functional from reduced density-matrix representations}},
author = {Kirill Kulaev and Alexander Ryabov and Bogdan Protsenko and Alexander Guda and Alexander Soldatov and Evgeny Burnaev and Vladimir Vanovskiy},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.00826},
eprint = {2611.00826},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.00826},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - ROSE: Deep Learning exchange-correlation functional from reduced density-matrix representations
AU - Kirill Kulaev
AU - Alexander Ryabov
AU - Bogdan Protsenko
AU - Alexander Guda
AU - Alexander Soldatov
AU - Evgeny Burnaev
AU - Vladimir Vanovskiy
PY - 2026
DA - 2026/11/01
PB - arXiv
DO - 10.48550/arxiv.2611.00826
UR - https://arxiv.org/abs/2611.00826
AB - We introduce ROSE, a deep-learning exchange--correlation energy functional built from the occupied-state density matrix. On GMTKN55 and on additional external benchmarks considered in this work, ROSE improves over modern neural network functionals trained under matched protocols and is on average more accurate than, or on par with, most hybrid and range-separated hybrid functionals considered here. On the experimental benchmarks considered here it has leading accuracy across hybrids, range separated hybrids, and neural functionals. These results motivate nonlocal density-matrix representations as a route to systematic improvement of learned XC functionals with low computational cost.
ER -