Volume 4, Issue 2
BINet: Learn to Solve Partial Differential Equations with Boundary Integral Networks

Guochang Lin, Pipi Hu, Fukai Chen, Xiang Chen, Junqing Chen, Jun Wang & Zuoqiang Shi

CSIAM Trans. Appl. Math., 4 (2023), pp. 275-305.

Published online: 2023-02

Export citation
  • Abstract

We propose a method combining boundary integral equations and neural networks (BINet) to solve (parametric) partial differential equations (PDEs) and operator problems in both bounded and unbounded domains. For PDEs with explicit fundamental solutions, BINet learns to solve, as a proxy, associated boundary integral equations using neural networks. The benefits are three-fold. Firstly, only the boundary conditions need to be fitted since the PDE can be automatically satisfied with single or double layer potential according to the potential theory. Secondly, the dimension of the boundary integral equations is less by one, and as such, the sample complexity can be reduced significantly. Lastly, in the proposed method, all differential operators have been removed, hence the numerical efficiency and stability are improved. Adopting neural tangent kernel (NTK) techniques, we provide proof of the convergence of BINets in the limit that the width of the neural network goes to infinity. Extensive numerical experiments show that, without calculating high-order derivatives, BINet is much easier to train and usually gives more accurate solutions, especially in the cases that the boundary conditions are not smooth enough. Further, BINet outperforms strong baselines for both one single PDE and parameterized PDEs in the bounded and unbounded domains.

  • AMS Subject Headings

65N35, 65N80, 68T20

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{CSIAM-AM-4-275, author = {Lin , GuochangHu , PipiChen , FukaiChen , XiangChen , JunqingWang , Jun and Shi , Zuoqiang}, title = {BINet: Learn to Solve Partial Differential Equations with Boundary Integral Networks}, journal = {CSIAM Transactions on Applied Mathematics}, year = {2023}, volume = {4}, number = {2}, pages = {275--305}, abstract = {

We propose a method combining boundary integral equations and neural networks (BINet) to solve (parametric) partial differential equations (PDEs) and operator problems in both bounded and unbounded domains. For PDEs with explicit fundamental solutions, BINet learns to solve, as a proxy, associated boundary integral equations using neural networks. The benefits are three-fold. Firstly, only the boundary conditions need to be fitted since the PDE can be automatically satisfied with single or double layer potential according to the potential theory. Secondly, the dimension of the boundary integral equations is less by one, and as such, the sample complexity can be reduced significantly. Lastly, in the proposed method, all differential operators have been removed, hence the numerical efficiency and stability are improved. Adopting neural tangent kernel (NTK) techniques, we provide proof of the convergence of BINets in the limit that the width of the neural network goes to infinity. Extensive numerical experiments show that, without calculating high-order derivatives, BINet is much easier to train and usually gives more accurate solutions, especially in the cases that the boundary conditions are not smooth enough. Further, BINet outperforms strong baselines for both one single PDE and parameterized PDEs in the bounded and unbounded domains.

}, issn = {2708-0579}, doi = {https://doi.org/10.4208/csiam-am.SO-2022-0014}, url = {http://global-sci.org/intro/article_detail/csiam-am/21415.html} }
TY - JOUR T1 - BINet: Learn to Solve Partial Differential Equations with Boundary Integral Networks AU - Lin , Guochang AU - Hu , Pipi AU - Chen , Fukai AU - Chen , Xiang AU - Chen , Junqing AU - Wang , Jun AU - Shi , Zuoqiang JO - CSIAM Transactions on Applied Mathematics VL - 2 SP - 275 EP - 305 PY - 2023 DA - 2023/02 SN - 4 DO - http://doi.org/10.4208/csiam-am.SO-2022-0014 UR - https://global-sci.org/intro/article_detail/csiam-am/21415.html KW - Partial differential equation, boundary integral, neural network, learning operator. AB -

We propose a method combining boundary integral equations and neural networks (BINet) to solve (parametric) partial differential equations (PDEs) and operator problems in both bounded and unbounded domains. For PDEs with explicit fundamental solutions, BINet learns to solve, as a proxy, associated boundary integral equations using neural networks. The benefits are three-fold. Firstly, only the boundary conditions need to be fitted since the PDE can be automatically satisfied with single or double layer potential according to the potential theory. Secondly, the dimension of the boundary integral equations is less by one, and as such, the sample complexity can be reduced significantly. Lastly, in the proposed method, all differential operators have been removed, hence the numerical efficiency and stability are improved. Adopting neural tangent kernel (NTK) techniques, we provide proof of the convergence of BINets in the limit that the width of the neural network goes to infinity. Extensive numerical experiments show that, without calculating high-order derivatives, BINet is much easier to train and usually gives more accurate solutions, especially in the cases that the boundary conditions are not smooth enough. Further, BINet outperforms strong baselines for both one single PDE and parameterized PDEs in the bounded and unbounded domains.

Lin , GuochangHu , PipiChen , FukaiChen , XiangChen , JunqingWang , Jun and Shi , Zuoqiang. (2023). BINet: Learn to Solve Partial Differential Equations with Boundary Integral Networks. CSIAM Transactions on Applied Mathematics. 4 (2). 275-305. doi:10.4208/csiam-am.SO-2022-0014
Copy to clipboard
The citation has been copied to your clipboard