Softmax parameterization of the occupation numbers for natural orbital functionals based on electron pairing approaches

Abstract

Within the framework of natural orbital functional theory, having a convenient representation of the occupation numbers and orbitals becomes critical for the computational performance of the calculations. Recognizing this, we propose an innovative parametrization of the occupation numbers that takes advantage of the electron-pairing approach used in Piris natural orbital functionals through the adoption of the softmax function, a pivotal component in modern deep-learning models. Our approach not only ensures adherence to the N-representability of the first-order reduced density matrix (1RDM) but also significantly enhances the computational efficiency of 1RDM functional theory calculations. The effectiveness of this alternative parameterization approach was assessed using the W4-17-MR molecular set, which demonstrated faster and more robust convergence compared to previous implementations.

13 pages, 6 Figures

References in corpus (21)

• A natural orbital functional for the many-electron problem
• Global Method for Electron Correlation
• Benchmark calculations for reduced density-matrix functional theory
• Global Natural Orbital Functional: Towards the Complete Description of the Electron Correlation
• Relating the pure and ensemble density matrix functional
• Diverging exchange force and form of the exact density matrix functional
• Ensemble reduced density matrix functional theory for excited states and hierarchical generalization of Pauli's exclusion principle
• DoNOF: an open-source implementation of natural-orbital-functional-based methods for quantum chemistry
• Foundation of one-particle reduced density matrix functional theory for excited states
• Natural orbital functional for multiplets
• Refining and relating fundamentals of functional theory
• Electron correlation in the Iron(II) Porphyrin by NOF approximations
• Exploiting the hessian for a better convergence of the SCF RDMFT procedure
• Reducing the Quantum Many-electron Problem to Two Electrons with Machine Learning
• Inverse molecular design and parameter optimization with Hückel theory using automatic differentiation
• Three Decades of Activations: A Comprehensive Survey of 400 Activation Functions for Neural Networks
• Outstanding Improvement in Removing the Delocalization Error by Global Natural Orbital Functional
• Data-driven Refinement of Electronic Energies from Two-Electron Reduced-Density-Matrix Theory
• 1-matrix functional for long-range interaction energy of two hydrogen atoms
• $\Delta$NO and the complexities of electron correlation in simple hydrogen clusters
• Enhancing Reduced Density Matrix Functional Theory Calculations by Coupling Orbital and Occupation Optimizations