.png){kind=icon}
.png){kind=icon}
Return
Physics-informed convolution gated recurrent unit network for solving an inverse problem
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neucom.2024.128254.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Deep learning has gained considerable attention in recent years for solving partial differential equations (PDEs) and inverse problems. In this paper a novel framework called physics-informed convolution gated recurrent unit Resnet (PhyCGRUR) is introduced. It is designed to address PDEs and their associated inverse problems. To investigate our model, first we apply the PhyCGRUR for solving PDEs without any labeled data. Then, we adapt the proposed methodology for identifying some parameters in PDEs from the final observation. Finally, we illustrate the capabilities and robustness of our proposed architecture (PhyCGRUR) compared with some competitive learned models using relative error, complexity, memory requirements and Running Time.
Keywords:
Neural network
Deep learning
Inverse problem
PhyCGRUR architecture
PDE
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W