arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2505.03213physics.chem-phquant-ph
收藏

迈向改进的量子机器学习用于分子力场

Towards Improved Quantum Machine Learning for Molecular Force Fields

The University of Tokyo(东京大学);Quemix Inc.(Quemix公司);National Institutes for Quantum Science and Technology (QST)(量子科学技术研究机构);NGK INSULATORS, LTD.(NGK绝缘体有限公司)

Yannick Couzinié, Shunsuke Daimon, Hirofumi Nishi, Natsuki Ito, Yusuke Harazono, Yu-ichiro Matsushita

AI总结:

本文提出改进的量子神经网络架构以提升分子力场生成的准确性,但发现其在能量预测和误差缩放方面存在局限。

摘要:

本研究探讨了使用等变量子神经网络(QNN)生成分子力场的应用,重点研究rMD17数据集。我们考虑了一种基于先前研究的QNN架构,并指出了原子环境参数化方面的不足。这些不足限制了其作为原子间势能的表达能力,并阻碍了分子间的可转移性。我们提出了一种改进的QNN架构,以解决这些不足。尽管两种QNN在力预测方面都显示出前景,但改进的架构在准确性上有所提升,但两者在能量预测上都表现出困难。此外,两种QNN架构未能展示出随着训练数据增加而误差减少的有意义的缩放定律。这些发现突显了将QNN扩展到复杂分子系统中的挑战,并强调了改进的编码策略、正则化技术以及混合量子-经典方法的必要性。

英文摘要:

This study explores the use of equivariant quantum neural networks (QNN) for generating molecular force fields, focusing on the rMD17 dataset. We consider a QNN architecture based on previous research and point out shortcomings in the parametrization of the atomic environments. These shortcomings limit its expressivity as an interatomic potential and precludes transferability between molecules. We propose a revised QNN architecture that addresses these shortcomings. While both QNNs show promise in force prediction, with the revised architecture showing improved accuracy, they struggle with energy prediction. Further, both QNNs architectures fail to demonstrate a meaningful scaling law of decreasing errors with increasing training data. These findings highlight the challenges of scaling QNNs for complex molecular systems and emphasize the need for improved encoding strategies, regularization techniques, and hybrid quantum-classical approaches.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑