2025-09-19 ミシガン大学
A 3D map of the quantum potential that guides the positions and motions of electrons in lithium hydride. The purple region is least favorable, while the orange is most favorable. The cut-away section shows the centers of the lithium (left) and hydrogen (right) atoms. Image credit: Bikash Kanungo and Paul Zimmerman, University of Michigan
＜関連情報＞
- https://news.umich.edu/quantum-chemistry-making-key-simulation-approach-more-accurate/
- https://www.science.org/doi/10.1126/sciadv.ady8962
正確な交換相関ポテンシャルとエネルギーから局所および半局所密度汎関数を学ぶ Learning local and semi-local density functionals from exact exchange-correlation potentials and energies
Bikash Kanungo, Jeffrey Hatch, Paul M. Zimmerman, and Vikram Gavini
Science Advances Published:19 Sep 2025
DOI:https://doi.org/10.1126/sciadv.ady8962
Abstract
Finding accurate exchange-correlation (XC) functionals remains the defining challenge in density functional theory (DFT). Despite 40 years of active development, attaining general purpose chemical accuracy is still elusive with existing functionals. We present a data-driven pathway to learn the XC functional by using the exact density, XC energy, and XC potential. While the exact densities are obtained from accurate configuration interaction (CI), the exact XC energies and XC potentials are obtained via inverse DFT calculations on the CI densities. We demonstrate how simple neural network (NN)–based local density approximation (LDA) and generalized gradient approximation (GGA), trained on just five atoms and two molecules, provide remarkable improvement in total energies and densities. Particularly, the NN-based GGA functional attains similar accuracy as the higher rung SCAN meta-GGA on various thermochemistry datasets. These results underscore the promise of using the XC potential in modeling XC functionals and can pave the way for systematic learning of increasingly accurate XC functionals.
