Machine learning framework for GW calculations across molecular dynamics trajectories
學年 114
學期 2
出版（發表）日期 2026-07-02
作品名稱 Machine learning framework for GW calculations across molecular dynamics trajectories
作品名稱（其他語言）
著者 Ragab A Abdelghany; Chih-En Hsu; Hung-Chung Hsueh; Yuan-Hong Tsai; Ming-Chiang Chung
單位
出版者
著錄名稱、卷期、頁數 Machine Learning: Science and Technology 7, p.045003
摘要 Machine learning offers a transformative approach to overcoming the computational bottlenecks of quasiparticle energy calculations in many-body perturbation theory. While the G0W0 approx￾imation provides a rigorous description of excited-state electronic properties, its high computa￾tional cost—driven by the evaluation of the inverse dielectric matrix—renders it prohibitive for molecular dynamics (MD) simulations that require sampling thousands of thermal configura￾tions. We present a gradient-boosting framework (LightGBM) that predicts G0W0 quasiparticle energies with high efficiency using only computationally inexpensive mean-field density functional theory eigenvalues and exchange-correlation potentials. Trained on a minimal subset (25%) of MD snapshots for silicon and boron nitride, the model achieves root-mean-square errors below 0.1 eV, accurately reproducing full k-resolved band structures and density of states. Crucially, we demon￾strate the model’s transferability and robustness: it successfully generalizes to unseen simulation times, different thermal regimes, and distinct crystal polymorphs excluded from the training data. This data-driven approach effectively bypasses the explicit calculation of the self-energy operator, offering a scalable pathway for high-throughput excited-state simulations and the study of finite￾temperature electronic properties in complex materials.
關鍵字
語言 en_US
ISSN 2632-2153
期刊性質 國外
收錄於 SCI
產學合作
通訊作者 Hung-Chung Hsueh, Yuan-Hong Tsai, Ming-Chiang Chung
審稿制度 是
國別 USA
公開徵稿
出版型式 ,電子版
相關連結

機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/129742 )

SDGS 優質教育