Fast and Scalable Neural Network Quantum States Method for Molecular Potential Energy Surfaces

doi:10.1109/TPDS.2025.3568360

	Fast and Scalable Neural Network Quantum States Method for Molecular Potential Energy Surfaces
	Wu, Yangjun 1; Cao, Wanlu 2; Zhao, Jiacheng 2; Shang, Honghui 1
	2025-07-01
发表期刊	IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
ISSN	1045-9219
卷号	36 期号:7 页码:1431-1443
摘要	The Neural Network Quantum States (NNQS) method is highly promising for accurately solving the Schr & ouml;dinger equation, yet it encounters challenges such as computational demands and slow rates of convergence. To address the high computational requirements, we introduce optimizations including a cross-sample KV cache sharing technique to enhance sampling efficiency, Quantum Bitwise and BloomHash methods for more efficient local energy computation, and mixed-precision training strategies to boost computational efficiency. To overcome the issue of slow convergence, we propose a parallel training algorithm for NNQS under second quantization to accelerate the training of base models for molecular potential surfaces. Our approach achieves up to 27-fold acceleration specifically in local energy calculations in systems with 154 spin orbitals and demonstrates strong and weak scaling efficiencies of 98% and 97%, respectively, on the H$_{2}$2O$_{2}$2 potential surface training set. The parallelized implementation of transformer-based NNQS is highly portable on various high-performance computing architectures, offering new perspectives on quantum chemistry simulations.
关键词	Artificial neural networks Computational efficiency Training Wave functions Quantum state Computational modeling Optimization Electrons Convergence Potential energy Quantum computational chemistry many-body Schr & ouml neural network quantum state transformer based architecture autoregressive sampling potential energy surfaces dinger equation
DOI	10.1109/TPDS.2025.3568360
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[T2222026] ; National Natural Science Foundation of China[U23B2020] ; National Natural Science Foundation of China[62302479] ; National Natural Science Foundation of China[62232015]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS记录号	WOS:001498254200002
出版者	IEEE COMPUTER SOC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/42312
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Shang, Honghui
作者单位	1.Univ Sci & Technol China, State Key Lab Precis & Intelligent Chem, Hefei 230026, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Wu, Yangjun,Cao, Wanlu,Zhao, Jiacheng,et al. Fast and Scalable Neural Network Quantum States Method for Molecular Potential Energy Surfaces[J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,2025,36(7):1431-1443.
APA	Wu, Yangjun,Cao, Wanlu,Zhao, Jiacheng,&Shang, Honghui.(2025).Fast and Scalable Neural Network Quantum States Method for Molecular Potential Energy Surfaces.IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,36(7),1431-1443.
MLA	Wu, Yangjun,et al."Fast and Scalable Neural Network Quantum States Method for Molecular Potential Energy Surfaces".IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 36.7(2025):1431-1443.