Fitting potential energy surfaces with fundamental invariant neural network. II. Generating fundamental invariants for molecular systems with up to ten atoms

被引:68
|
作者
Chen, Rongjun [1 ,2 ,3 ]
Shao, Kejie [1 ,2 ]
Fu, Bina [1 ,2 ]
Zhang, Dong H. [1 ,2 ]
机构
[1] Chinese Acad Sci, State Key Lab Mol React Dynam, Dalian Inst Chem Phys, Dalian 116023, Peoples R China
[2] Chinese Acad Sci, Ctr Theoret & Computat Chem, Dalian Inst Chem Phys, Dalian 116023, Peoples R China
[3] Univ Chinese Acad Sci, Beijing 100049, Peoples R China
来源
JOURNAL OF CHEMICAL PHYSICS | 2020年 / 152卷 / 20期
基金
中国国家自然科学基金; 国家重点研发计划;
关键词
DYNAMICS; REGRESSION;
D O I
10.1063/5.0010104
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
Symmetry adaptation is crucial in representing a permutationally invariant potential energy surface (PES). Due to the rapid increase in computational time with respect to the molecular size, as well as the reliance on the algebra software, the previous neural network (NN) fitting with inputs of fundamental invariants (FIs) has practical limits. Here, we report an improved and efficient generation scheme of FIs based on the computational invariant theory and parallel program, which can be readily used as the input vector of NNs in fitting high-dimensional PESs with permutation symmetry. The newly developed method significantly reduces the evaluation time of FIs, thereby extending the FI-NN method for constructing highly accurate PESs to larger systems beyond five atoms. Because of the minimum size of invariants used in the inputs of the NN, the NN structure can be very flexible for FI-NN, which leads to small fitting errors. The resulting FI-NN PES is much faster on evaluating than the corresponding permutationally invariant polynomial-NN PES.
引用
收藏
页数:8
相关论文
共 13 条
  • [1] Communication: Fitting potential energy surfaces with fundamental invariant neural network
    Shao, Kejie
    Chen, Jun
    Zhao, Zhiqiang
    Zhang, Dong H.
    JOURNAL OF CHEMICAL PHYSICS, 2016, 145 (07):
  • [2] Permutation invariant polynomial neural network approach to fitting potential energy surfaces. II. Four-atom systems
    Li, Jun
    Jiang, Bin
    Guo, Hua
    JOURNAL OF CHEMICAL PHYSICS, 2013, 139 (20):
  • [3] Accurate fundamental invariant-neural network representation of ab initio potential energy surfaces
    Fu, Bina
    Zhang, Dong H.
    NATIONAL SCIENCE REVIEW, 2023, 10 (12)
  • [4] Accurate fundamental invariant-neural network representation of ab initio potential energy surfaces
    Bina Fu
    Dong H.Zhang
    National Science Review, 2023, 10 (12) : 26 - 39
  • [5] Permutation invariant polynomial neural network approach to fitting potential energy surfaces
    Jiang, Bin
    Guo, Hua
    JOURNAL OF CHEMICAL PHYSICS, 2013, 139 (05):
  • [6] Permutation invariant polynomial neural network approach to fitting potential energy surfaces. IV. Coupled diabatic potential energy matrices
    Xie, Changjian
    Zhu, Xiaolei
    Yarkony, David R.
    Guo, Hua
    JOURNAL OF CHEMICAL PHYSICS, 2018, 149 (14):
  • [7] The full-dimensional potential energy surface of He2H- using fundamental invariant neural network method
    Bai, Yuyao
    Fu, Yan-Lin
    Wang, Gao-Ren
    Fu, Bina
    Han, Yong-Chang
    INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY, 2022, 122 (22)
  • [8] Potential energy surfaces from high fidelity fitting of ab initio points: the permutation invariant polynomial - neural network approach
    Jiang, Bin
    Li, Jun
    Guo, Hua
    INTERNATIONAL REVIEWS IN PHYSICAL CHEMISTRY, 2016, 35 (03) : 479 - 506
  • [9] Fundamental invariant-neural network potential energy surface and dissociative chemisorption dynamics of N2 on rigid Ni(111)
    Shi, Huixia
    Liu, Tianhui
    Fu, Yanlin
    Wu, Hao
    Fu, Bina
    Zhang, Dong H.
    COMPUTATIONAL AND THEORETICAL CHEMISTRY, 2022, 1211
  • [10] Roaming Dynamics of H+C2D2 Reaction on Fundamental-Invariant Neural Network Potential Energy Surface
    Bai, Yuyao
    Fu, Yan-Lin
    Han, Yong-Chang
    Fu, Bina
    Zhang, Dong H.
    CHINESE JOURNAL OF CHEMICAL PHYSICS, 2022, 35 (02) : 295 - 302