Scalable neural quantum states architecture for quantum chemistry

Scalable neural quantum states architecture for quantum chemistry
复制标题

用于量子化学的可扩展神经量子态架构

DOI:
10.1088/2632-2153/acdb2f
复制
发表时间:
2023
期刊:
Machine Learning: Science and Technology
影响因子:
--
通讯作者:
Veerapaneni, Shravan
Veerapaneni, Shravan
中科院分区:
--
文献类型:
--
作者:
Zhao, Tianchen;Stokes, James;Veerapaneni, Shravan

文献摘要

参考文献

被引文献

相似文献

量子态的神经网络表示的变分优化已成功地应用于求解相互作用费米子问题。尽管发展迅速,但在考虑大尺度分子时,会出现重大的可扩展性挑战,这些分子对应于由数千甚至数百万泡利算子组成的非局部相互作用量子自旋哈密顿量。在这项工作中,我们引入了可扩展的并行化策略,以改进基于神经网络的变分量子蒙特卡罗计算,用于从头算量子化学应用。我们建立了gpu支持的局部能量并行来计算潜在复杂分子的哈密顿量的优化目标。使用自回归采样技术,我们演示了系统改进壁钟时间,以实现具有高达双激发基线目标能量的耦合簇。通过将自旋哈密顿量的结构调整到自回归采样顺序中,进一步提高了性能。与经典的近似方法相比,该算法具有良好的性能,并且在运行时间和可扩展性方面都优于现有的基于神经网络的方法。
Variational optimization of neural-network representations of quantum states has been successfully applied to solve interacting fermionic problems. Despite rapid developments, significant scalability challenges arise when considering molecules of large scale, which correspond to non-locally interacting quantum spin Hamiltonians consisting of sums of thousands or even millions of Pauli operators. In this work, we introduce scalable parallelization strategies to improve neural-network-based variational quantum Monte Carlo calculations for ab-initio quantum chemistry applications. We establish GPU-supported local energy parallelism to compute the optimization objective for Hamiltonians of potentially complex molecules. Using autoregressive sampling techniques, we demonstrate systematic improvement in wall-clock timings required to achieve coupled cluster with up to double excitations baseline target energies. The performance is further enhanced by accommodating the structure of resultant spin Hamiltonians into the autoregressive sampling ordering. The algorithm achieves promising performance in comparison with the classical approximate methods and exhibits both running time and scalability advantages over existing neural-network based methods.
DOI: 10.1088/2058-9565/ab8ebc
发表时间: 2020-07-01
影响因子: 6.7
作者:
McClean, Jarrod R.;Rubin, Nicholas C.;Babbush, Ryan
通讯作者: Babbush, Ryan
DOI: 10.1126/science.aag2302
发表时间: 2017-02-10
期刊: SCIENCE
影响因子: 56.9
作者:
Carleo, Giuseppe;Troyer, Matthias
通讯作者: Troyer, Matthias
DOI: 10.1524/zpch.1995.192.part_2.221
发表时间: 1995
期刊: Zeitschrift für Physikalische Chemie
影响因子: --
作者:
K. Fiedler
通讯作者: K. Fiedler
从头算量子化学的自回归神经网络波函数
DOI: 10.1038/s42256-022-00461-z
发表时间: 2021
影响因子: 23.8
作者:
T. Barrett;A. Malyshev;A. Lvovsky
通讯作者: A. Lvovsky