Abstract:Local hybrid functionals (LHs) are a modern class of density functionals that use a real-space position-dependent admixture of exact exchange (EXX), governed by a local mixing function (LMF). While many model LMFs have been proposed and evaluated over the past 10-20 years, their systematic construction has been hampered by a lack of exact physical constraints on their valence behavior. Here we exploit a data-driven approach and train a new type of "n-LMF" as a relatively shallow neural network. The input features of this n-LMF are of meta-GGA character, while the W4-17 atomization-energy and BH76 reaction-barrier test sets have been used for training. Simply replacing the widely used "t-LMF" of the LH20t functional by the n-LMF provides the LH24n-B95 functional. Augmented by DFT-D4 dispersion corrections, LH24n-B95-D4 remarkably improves the WTMAD-2 value for the large GMTKN55 test suite of general main-group thermochemistry, kinetics and noncovalent interactions (NCIs) from 4.55 kcal/mol to 3.49 kcal/mol. As we found the limited flexibility of the B95c correlation functional to disfavor much further improvement on NCIs, we proceeded to replace it by an optimized B97c-type power-series expansion. This gives the LH24n functional. LH24n-D4 gives a WTMAD-2 value of 3.10 kcal/mol, the so far lowest value of a rung 4 functional in self-consistent calculations. The new functionals perform moderately well for organometallic transition-metal energetics while leaving room for further data-driven improvements in that area. Compared to complete neural-network functionals like DM21, the present more tailored approach to train just the LMF in a flexible but well-defined human-designed LH functional retains the possibility of graphical LMF analyses to gain deeper understanding. We find that both the present n-LMF and the recent x-LMF suppress the so-called gauge problem of local hybrids without adding a calibration function as required for other LMFs like the t-LMF. LMF plots show that this can be traced back to large LMF values in the small-density region between the interacting atoms in NCIs for n- and x-LMFs and low values for the t-LMF. We also find that the trained n-LMF has relatively large values in covalent bonds without deteriorating binding energies. The current approach enables fast and efficient routine self-consistent calculations using n-LMFs in Turbomole. Further routes toward improved functionals are delineated.

A generalized exchange-correlation functional: the Neural-Networks approach

Improving the Performance of Long-Range-Corrected Exchange-Correlation Functional with an Embedded Neural Network

Consistent Improvement on Both Exchange and Correlation within Hybrid Multiconfiguration Pair-density Functional Theory: tB4LYP

Modeling Intermolecular Interactions with Exchange-Hole Dipole Moment Dispersion Corrections to Neural Network Potentials

Performance of a nonempirical exchange functional from the density matrix expansion: comparative study with different correlation

Learning local and semi-local density functionals from exact exchange-correlation potentials and energies

Calculate electronic excited states using neural networks with effective core potential

Improving the Accuracy of Density Functional Theory (DFT) Calculation for Homolysis Bond Dissociation Energies of Y-NO Bond: Generalized Regression Neural Network Based on Grey Relational Analysis and Principal Component Analysis

Numerical Examination of Several Exchange Energy Functionals

Introducing a new exchange functional by altering the electron density's ionization dependency in density functional theory

A long-range correction scheme for generalized-gradient-approximation exchange functionals

Improving the B 3 LYP bond energies by using the X 1 method

An extended hybrid density functional (X3LYP) with improved descriptions of nonbond interactions and thermodynamic properties of molecular systems.

An Artificial Neural Network-based Density Functional Approach for Adiabatic Energy Differences in Transition Metal Complexes

Quantum-Enhanced Neural Exchange-Correlation Functionals

Data-Driven Improvement of Local Hybrid Functionals: Neural-Network-Based Local Mixing Functions and Power-Series Correlation Functionals

Atoms, molecules, solids, and surfaces: Applications of the generalized gradient approximation for exchange and correlation.

Improving the B3lyp Bond Energies by Using the X1 Method

Enforcing the Linear Behavior of the Total Energy with Hybrid Functionals: Implications for Charge Transfer, Interaction Energies, and the Random-Phase Approximation

Ab Initio Calculation of Real Solids Via Neural Network Ansatz

Modeling Intermolecular Interactions With XDM Dispersion Corrections to Neural Network Potentials