An unfitted RBF-FD method in a least-squares setting for elliptic PDEs on complex geometries.
Igor TominecEva BreznikPublished in: J. Comput. Phys. (2021)
Keyphrases
- least squares
- high accuracy
- objective function
- artificial neural networks
- support vector machine svm
- support vector machine
- regularization parameter
- neural network
- rbf network
- rbf neural network
- sensitivity analysis
- clustering method
- parameter estimation
- linear combination
- probabilistic model
- computational cost
- dynamic programming
- cost function
- significant improvement
- similarity measure
- image processing
- learning algorithm