A neural network training algorithm for singular perturbation boundary value problems.
T. E. SimosIoannis Th. FamelisPublished in: Neural Comput. Appl. (2022)
Keyphrases
- training algorithm
- neural network
- boundary value problem
- back propagation
- boundary conditions
- neural network training
- differential equations
- training process
- partial differential equations
- artificial neural networks
- numerical solution
- gradient vector
- variational inequalities
- feed forward
- rbf neural network
- hidden layer
- learning rate
- learning algorithm
- support vector machine
- data sets
- radial basis function
- training data
- activation function
- rbf network
- fuzzy logic
- multilayer perceptron
- dynamical systems
- multiscale
- sufficient conditions
- decision trees
- shape from shading