Separable graph Hamiltonian network: A graph deep learning model for lattice systems
Addressing the challenges posed by nonlinear lattice models, which are vital across diverse scientific disciplines, we present a new deep learning approach that harnesses the power of graph neural networks. By representing the lattice system as a graph and leveraging the graph structures to identify...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
American Physical Society
2024-02-01
|
| Series: | Physical Review Research |
| Online Access: | http://doi.org/10.1103/PhysRevResearch.6.013176 |
| _version_ | 1797210228478967808 |
|---|---|
| author | Ru Geng Jian Zu Yixian Gao Hong-Kun Zhang |
| author_facet | Ru Geng Jian Zu Yixian Gao Hong-Kun Zhang |
| author_sort | Ru Geng |
| collection | DOAJ |
| description | Addressing the challenges posed by nonlinear lattice models, which are vital across diverse scientific disciplines, we present a new deep learning approach that harnesses the power of graph neural networks. By representing the lattice system as a graph and leveraging the graph structures to identify complex nonlinear relationships, we have developed a flexible solution that outperforms traditional techniques. Our model not only offers precise trajectory predictions and energy conservation properties by incorporating separable Hamiltonians but also proves superior to existing top-tier models when tested on classic nonlinear oscillator lattice problems: a mixed Fermi-Pasta-Ulam Klein-Gordon, a Klein-Gordon system with long-range interactions, and a two-dimensional Frenkel-Kontorova, highlighting its potential for wide-reaching applications. |
| first_indexed | 2024-04-24T10:07:16Z |
| format | Article |
| id | doaj.art-501c02c069344432b08e3e3c71e3b7ef |
| institution | Directory Open Access Journal |
| issn | 2643-1564 |
| language | English |
| last_indexed | 2024-04-24T10:07:16Z |
| publishDate | 2024-02-01 |
| publisher | American Physical Society |
| record_format | Article |
| series | Physical Review Research |
| spelling | doaj.art-501c02c069344432b08e3e3c71e3b7ef2024-04-12T17:39:16ZengAmerican Physical SocietyPhysical Review Research2643-15642024-02-016101317610.1103/PhysRevResearch.6.013176Separable graph Hamiltonian network: A graph deep learning model for lattice systemsRu GengJian ZuYixian GaoHong-Kun ZhangAddressing the challenges posed by nonlinear lattice models, which are vital across diverse scientific disciplines, we present a new deep learning approach that harnesses the power of graph neural networks. By representing the lattice system as a graph and leveraging the graph structures to identify complex nonlinear relationships, we have developed a flexible solution that outperforms traditional techniques. Our model not only offers precise trajectory predictions and energy conservation properties by incorporating separable Hamiltonians but also proves superior to existing top-tier models when tested on classic nonlinear oscillator lattice problems: a mixed Fermi-Pasta-Ulam Klein-Gordon, a Klein-Gordon system with long-range interactions, and a two-dimensional Frenkel-Kontorova, highlighting its potential for wide-reaching applications.http://doi.org/10.1103/PhysRevResearch.6.013176 |
| spellingShingle | Ru Geng Jian Zu Yixian Gao Hong-Kun Zhang Separable graph Hamiltonian network: A graph deep learning model for lattice systems Physical Review Research |
| title | Separable graph Hamiltonian network: A graph deep learning model for lattice systems |
| title_full | Separable graph Hamiltonian network: A graph deep learning model for lattice systems |
| title_fullStr | Separable graph Hamiltonian network: A graph deep learning model for lattice systems |
| title_full_unstemmed | Separable graph Hamiltonian network: A graph deep learning model for lattice systems |
| title_short | Separable graph Hamiltonian network: A graph deep learning model for lattice systems |
| title_sort | separable graph hamiltonian network a graph deep learning model for lattice systems |
| url | http://doi.org/10.1103/PhysRevResearch.6.013176 |
| work_keys_str_mv | AT rugeng separablegraphhamiltoniannetworkagraphdeeplearningmodelforlatticesystems AT jianzu separablegraphhamiltoniannetworkagraphdeeplearningmodelforlatticesystems AT yixiangao separablegraphhamiltoniannetworkagraphdeeplearningmodelforlatticesystems AT hongkunzhang separablegraphhamiltoniannetworkagraphdeeplearningmodelforlatticesystems |