A generalized exchange-correlation functional: the Neural-Networks approach ☆
By Xiao Zheng, Lihong Hu, X. Wang, Guanhua Chen. 2003
54 citations recorded by the source corpus.
A Neural-Networks approach is employed to improve B3LYP exchange-correlation functional by taking into account of high-order contributions. The new B3LYP functional is based on a Neural-Network whose structure and synaptic weights are determined from 116 known experimental energy data [J. Chem. Phys. 98 (1993) 5648]. It leads to better agreement between the first-principles calculations and the experimental results. The new functional is further tested by applying it to calculate 40 ionization potentials and 40 enthalpies of formation in G2-2 test set [J. Chem. Phys. 109 (1998) 42] using 6-311+G(3df,2p) basis set. The root-mean-square errors are reduced from those of conventional B3LYP calculations.
Subfields: Atomic and Molecular Physics, and Optics, Organic Chemistry
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Atomic and Molecular Physics, and Optics
- Patrick Rinke
- Matthias Scheffler
- Christoph Freysoldt
- Jörg Neugebauer
- Alexandre Tkatchenko
- O. A. von Lilienfeld
In Organic Chemistry
Citations in the corpus
- Combining GW calculations with exact-exchange density-functional theory: an analysis of valence-band photoemission for compound semiconductors (2005, methodology): “Examples include QSPRs for exchange-correlation potentials using neural networks (NNs) [2, 3], basisset effects using support vector machines [4, 5] or molecular reorganization energies affecting charge transfer rates [6, 7], or for solid ternary oxides [8].”