Guardado en:
Detalles Bibliográficos
Autores principales: Csirik, Mihály A., Laestadius, Andre, Oster, Mathias
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2511.18512
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917100103139328
author Csirik, Mihály A.
Laestadius, Andre
Oster, Mathias
author_facet Csirik, Mihály A.
Laestadius, Andre
Oster, Mathias
contents Density functional theory is one of the most efficient and widely used computational methods of quantum mechanics, especially in fields such as solid state physics and quantum chemistry. From the theoretical perspecive, its central object is the universal density functional which contains all intrinsic information about the quantum system in question. Once the external potential is provided, in principle one can obtain the exact ground-state energy via a simple minimization. However, the universal density functional is a very complicated mathematical object and almost always it is replaced with its approximate variants. So far, no ``first principles'', mathematically consistent and convergent approximation procedure has been devised that has general applicability. In this paper, we propose such a procedure by first applying Moreau--Yosida regularization to make the exact functionals continuous (even differentiable) and then approximate the regularized functional by a neural network. The resulting neural network preserves the positivity and convexity of the exact functionals. More importantly, it is differentiable, so it can be directly used in a Kohn--Sham calculation.
format Preprint
id arxiv_https___arxiv_org_abs_2511_18512
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Neural network approximation of regularized density functionals
Csirik, Mihály A.
Laestadius, Andre
Oster, Mathias
Chemical Physics
Disordered Systems and Neural Networks
Numerical Analysis
Mathematical Physics
Quantum Physics
81Q05, 68T07, 41A65
Density functional theory is one of the most efficient and widely used computational methods of quantum mechanics, especially in fields such as solid state physics and quantum chemistry. From the theoretical perspecive, its central object is the universal density functional which contains all intrinsic information about the quantum system in question. Once the external potential is provided, in principle one can obtain the exact ground-state energy via a simple minimization. However, the universal density functional is a very complicated mathematical object and almost always it is replaced with its approximate variants. So far, no ``first principles'', mathematically consistent and convergent approximation procedure has been devised that has general applicability. In this paper, we propose such a procedure by first applying Moreau--Yosida regularization to make the exact functionals continuous (even differentiable) and then approximate the regularized functional by a neural network. The resulting neural network preserves the positivity and convexity of the exact functionals. More importantly, it is differentiable, so it can be directly used in a Kohn--Sham calculation.
title Neural network approximation of regularized density functionals
topic Chemical Physics
Disordered Systems and Neural Networks
Numerical Analysis
Mathematical Physics
Quantum Physics
81Q05, 68T07, 41A65
url https://arxiv.org/abs/2511.18512