A hybrid MGA-MSGD ANN training approach for approximate solution of linear elliptic PDEs.
Hamidreza DehghaniAndreas ZilianPublished in: Math. Comput. Simul. (2021)
Keyphrases
- partial differential equations
- artificial neural networks
- using artificial neural networks
- feed forward neural networks
- neural network
- numerical solution
- evolutionary artificial neural networks
- multi layer perceptron
- image denoising
- training phase
- feed forward
- linear svm
- training examples
- training samples
- supervised learning
- anisotropic diffusion
- closed form
- image enhancement
- training set
- numerical scheme
- difference equations
- genetic algorithm