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

Badrinarayanan, Abhiram; Davidović, Davor; Napoli, Edoardo Di; Novak, Jurica; Genovese, Luigi; Ramirez‐Hidalgo, Gustavo; Wu, Xinzhe (2026) Data‐Driven Spectral Prediction for Accelerating Large‐Scale Electronic Structure Calculations. Proceedings in Applied Mathematics and Mechanics, 26 (4). ISSN 1617-7061

PDF - Published Version - article
Available under License Creative Commons Attribution.
Download (2MB)

Official URL: http://doi.org/10.1002/pamm.70204

Abstract

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.

Item Type:

Article

Uncontrolled Keywords:

algorithm; artificial neural network; Chebyshev polynomials; computer science; curse of dimensionality; Random Forest; rational function

Subjects:

NATURAL SCIENCES > Mathematics
NATURAL SCIENCES > Chemistry

Divisions:

Center for Informatics and Computing

Projects:

Project title	Project leader	Project code	Project type
Learning Materials at eXascale	UNSPECIFIED	101118139	Program Digitalna Europa
Skalabilni algoritmi visokih performansi za buduće heterogene distribuirane računalne sustave-HybridScale	Davor Davidović	UIP-2020-02-4559	HRZZ

Depositing User:

Jurica Novak

Date Deposited:

06 Oct 2026 06:35

URI:

https://fulir.irb.hr:/id/eprint/12140

DOI:

10.1002/pamm.70204