Enregistré dans:
Détails bibliographiques
Auteurs principaux: Hou, Bowen, Wu, Jinyuan, Lee, Victor Chang, Guo, Jiaxuan, Liu, Luna Y., Qiu, Diana Y.
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2502.05635
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910819932962816
author Hou, Bowen
Wu, Jinyuan
Lee, Victor Chang
Guo, Jiaxuan
Liu, Luna Y.
Qiu, Diana Y.
author_facet Hou, Bowen
Wu, Jinyuan
Lee, Victor Chang
Guo, Jiaxuan
Liu, Luna Y.
Qiu, Diana Y.
contents Many-body electron-hole interactions are essential for understanding non-linear optical processes and ultrafast spectroscopy of materials. Recent first principles approaches based on nonequilibrium Green's function formalisms, such as the time-dependent adiabatic GW (TD-aGW) approach, can predict the nonequilibrium dynamics of excited states including electron-hole interactions. However, the high dimensionality of the electron-hole kernel poses significant computational challenges for scalability. Here, we develop a data-driven low-rank approximation for the electron-hole kernel, leveraging localized excitonic effects in the Hilbert space of crystalline systems. Through singular value decomposition (SVD) analysis, we show that the subspace of non-zero singular values, containing the key information of the electron-hole kernel, retains a small size even as the k-grid grows, ensuring computational feasibility with extremely dense k-grids for converged calculations. Utilizing this low-rank property, we achieve at least 95% compression of the kernel and an order-of-magnitude speedup of TD-aGW calculations. Our method, rooted in physical interpretability, outperforms existing machine learning approaches by avoiding intensive training processes and eliminating time-accumulated errors, providing a general framework for high-throughput, nonequilibrium simulation of light-driven dynamics in materials.
format Preprint
id arxiv_https___arxiv_org_abs_2502_05635
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Data-driven Low-rank Approximation for Electron-hole Kernel and Acceleration of Time-dependent GW Calculations
Hou, Bowen
Wu, Jinyuan
Lee, Victor Chang
Guo, Jiaxuan
Liu, Luna Y.
Qiu, Diana Y.
Computational Physics
Many-body electron-hole interactions are essential for understanding non-linear optical processes and ultrafast spectroscopy of materials. Recent first principles approaches based on nonequilibrium Green's function formalisms, such as the time-dependent adiabatic GW (TD-aGW) approach, can predict the nonequilibrium dynamics of excited states including electron-hole interactions. However, the high dimensionality of the electron-hole kernel poses significant computational challenges for scalability. Here, we develop a data-driven low-rank approximation for the electron-hole kernel, leveraging localized excitonic effects in the Hilbert space of crystalline systems. Through singular value decomposition (SVD) analysis, we show that the subspace of non-zero singular values, containing the key information of the electron-hole kernel, retains a small size even as the k-grid grows, ensuring computational feasibility with extremely dense k-grids for converged calculations. Utilizing this low-rank property, we achieve at least 95% compression of the kernel and an order-of-magnitude speedup of TD-aGW calculations. Our method, rooted in physical interpretability, outperforms existing machine learning approaches by avoiding intensive training processes and eliminating time-accumulated errors, providing a general framework for high-throughput, nonequilibrium simulation of light-driven dynamics in materials.
title Data-driven Low-rank Approximation for Electron-hole Kernel and Acceleration of Time-dependent GW Calculations
topic Computational Physics
url https://arxiv.org/abs/2502.05635