Zipeng An | Searcharxiv

arXiv2026

PyGSC: A Python tool for correcting Kohn-Sham orbital energies by mitigating the delocalization error of density functional approximations

Density functional approximations (DFAs) suffer from delocalization error, which limits their accuracy in predicting electron affinities (EAs), ionization potentials (IPs), and quasiparticle energies. In this work, we present a theoretical refinement of the quasiparticle energies from density functional theory (QE-DFT) method by improving the perturbative expression for the exchange-correlation potential, leading to a more consistent description of molecular systems. We further develop an open-source Python program, PyGSC, built upon the PySCF library, which implements the modified QE-DFT framework. Benchmark tests on main-group atoms and G2/97 molecules demonstrate that the modified QE-DFT method outperforms the original DFAs, with third-order corrections achieving mean absolute deviations below 0.3 eV for EA and IP predictions. Application to dipole-bound states of DNA/RNA nucleobases further validates the superiority of the QE-DFT approach over original DFAs, offering an efficient and accurate approach for predicting electronic properties in large molecular systems.

physics.chem-ph↗

arXiv2025

Mitigating error cancellation in density functional approximations via machine learning correction

The integration of machine learning (ML) with density functional theory has emerged as a promising strategy to enhance the accuracy of density functional methods. While practical implementations of density functional approximations (DFAs) often exploit error cancellation between chemical species to achieve high accuracy in thermochemical and kinetic energy predictions, this approach is inherently system-dependent, which severely limits the transferability of DFAs. To address this challenge, we develop a novel ML-based correction to the widely used B3LYP functional, directly targeting its deviations from the exact exchange-correlation functional. By utilizing highly accurate absolute energies as exclusive reference data, our approach eliminates the reliance on error cancellation. To optimize the ML model, we attribute errors to real-space pointwise contributions and design a double-cycle protocol that incorporates self-consistent-field calculations into the training workflow. Numerical tests demonstrate that the ML model, trained solely on absolute energies, improves the accuracy of calculated relative energies, demonstrating that robust DFAs can be constructed without resorting to error cancellation. Comprehensive benchmarks further show that our ML-corrected B3LYP functional significantly outperforms the original B3LYP across diverse thermochemical and kinetic energy calculations, offering a versatile and superior alternative for practical applications.

physics.chem-ph↗