Data-Driven Spectral Prediction for Accelerating Large-Scale Electronic Structure Calculations

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Badrinarayanan, Abhiram, Davidovic, Davor, Di Napoli, Edoardo, Novak, Jurica, Genovese, Luigi, Ramirez-Hidalgo, Gustavo, Wu, Xinzhe
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910273684635648
author Badrinarayanan, Abhiram
Davidovic, Davor
Di Napoli, Edoardo
Novak, Jurica
Genovese, Luigi
Ramirez-Hidalgo, Gustavo
Wu, Xinzhe
author_facet Badrinarayanan, Abhiram
Davidovic, Davor
Di Napoli, Edoardo
Novak, Jurica
Genovese, Luigi
Ramirez-Hidalgo, Gustavo
Wu, Xinzhe
contents Simulating large molecular systems comprising thousands of atoms requires highly scalable methodologies. While modern Density Functional Theory (DFT) codes exhibit linear scaling, solving the associated large, sparse generalized eigenproblems remains a critical computational bottleneck on exascale architectures. In the context of the LimitX project, we propose a data-driven framework to accelerate these calculations. By shifting the machine learning target from discrete eigenvalues to the coefficients of an interpolating Chebyshev polynomial, and by comparing both all-atom and fragment-based structural representations, we successfully overcome the dimensionality constraints of large-scale spectral prediction. We investigate three machine learning models (Kernel Ridge Regression, Graph Neural Networks, and Random Forests) trained on a novel 2 TB dataset of protein dimers. The predicted spectra provide initial guesses that effectively bypass early Self-Consistent Field (SCF) iterations in BigDFT. Ultimately, these spectral predictors will be deployed to dynamically optimize upcoming rational filter-based eigensolvers, such as FrASE, which is currently in initial development.
format Preprint
id arxiv_https___arxiv_org_abs_2606_00401
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Data-Driven Spectral Prediction for Accelerating Large-Scale Electronic Structure Calculations
Badrinarayanan, Abhiram
Davidovic, Davor
Di Napoli, Edoardo
Novak, Jurica
Genovese, Luigi
Ramirez-Hidalgo, Gustavo
Wu, Xinzhe
Computational Physics
Materials Science
Machine Learning
Numerical Analysis
Simulating large molecular systems comprising thousands of atoms requires highly scalable methodologies. While modern Density Functional Theory (DFT) codes exhibit linear scaling, solving the associated large, sparse generalized eigenproblems remains a critical computational bottleneck on exascale architectures. In the context of the LimitX project, we propose a data-driven framework to accelerate these calculations. By shifting the machine learning target from discrete eigenvalues to the coefficients of an interpolating Chebyshev polynomial, and by comparing both all-atom and fragment-based structural representations, we successfully overcome the dimensionality constraints of large-scale spectral prediction. We investigate three machine learning models (Kernel Ridge Regression, Graph Neural Networks, and Random Forests) trained on a novel 2 TB dataset of protein dimers. The predicted spectra provide initial guesses that effectively bypass early Self-Consistent Field (SCF) iterations in BigDFT. Ultimately, these spectral predictors will be deployed to dynamically optimize upcoming rational filter-based eigensolvers, such as FrASE, which is currently in initial development.
title Data-Driven Spectral Prediction for Accelerating Large-Scale Electronic Structure Calculations
topic Computational Physics
Materials Science
Machine Learning
Numerical Analysis
url https://arxiv.org/abs/2606.00401