Saved in:
Bibliographic Details
Main Author: Zhang, Xiaoyu
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.05345
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910106798522368
author Zhang, Xiaoyu
author_facet Zhang, Xiaoyu
contents Density functional theory (DFT) and linear-response time-dependent density functional theory (LR-TDDFT) rely on an exchange-correlation (xc) approximation that provides not only energy but also its functional derivatives that enter the self-consistent potential and the response kernel. Here, we present an end-to-end differentiable workflow to optimize a single deep-learned energy functional using targets from both Kohn-Sham DFT and adiabatic LR-TDDFT. To enable this training in a computationally efficient and differentiable manner, we developed a JAX-based two-component quantum chemistry framework (IQC), in which the learned functional provides a self-consistent potential and linear-response kernel via automatic differentiation. This construction permits gradient-based optimization through both the self-consistent-field (SCF) fixed-point equations and the Casida eigenvalue problem. We learn an exchange-correlation functional on excitation energies of small molecules while incorporating one-electron self-interaction cancelation as penalty terms, and we assess its possible transfer to molecular test cases.
format Preprint
id arxiv_https___arxiv_org_abs_2602_05345
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle End-to-End Differentiable Learning of a Single Functional for DFT and Linear-Response TDDFT
Zhang, Xiaoyu
Chemical Physics
Density functional theory (DFT) and linear-response time-dependent density functional theory (LR-TDDFT) rely on an exchange-correlation (xc) approximation that provides not only energy but also its functional derivatives that enter the self-consistent potential and the response kernel. Here, we present an end-to-end differentiable workflow to optimize a single deep-learned energy functional using targets from both Kohn-Sham DFT and adiabatic LR-TDDFT. To enable this training in a computationally efficient and differentiable manner, we developed a JAX-based two-component quantum chemistry framework (IQC), in which the learned functional provides a self-consistent potential and linear-response kernel via automatic differentiation. This construction permits gradient-based optimization through both the self-consistent-field (SCF) fixed-point equations and the Casida eigenvalue problem. We learn an exchange-correlation functional on excitation energies of small molecules while incorporating one-electron self-interaction cancelation as penalty terms, and we assess its possible transfer to molecular test cases.
title End-to-End Differentiable Learning of a Single Functional for DFT and Linear-Response TDDFT
topic Chemical Physics
url https://arxiv.org/abs/2602.05345