Physics-informed graph neural Galerkin networks: A unified framework for solving PDE-governed forward and inverse problems.
Han GaoMatthew J. ZahrJian-Xun WangPublished in: CoRR (2021)
Keyphrases
- partial differential equations
- inverse problems
- level set
- anisotropic diffusion
- image denoising
- image processing
- image enhancement
- regularization method
- energy functional
- image reconstruction
- emission tomography
- differential equations
- neural network
- multiscale
- optimization methods
- connected components
- global optimization
- image sequences
- denoising
- evolutionary algorithm
- weighted graph
- high order
- convex optimization
- computational complexity
- image segmentation
- genetic algorithm