A novel class of approximate inverse preconditioners for large positive definite linear systems in optimization.
Giovanni FasanoMassimo RomaPublished in: Comput. Optim. Appl. (2016)
Keyphrases
- linear systems
- positive definite
- sufficient conditions
- dynamical systems
- conjugate gradient method
- kernel function
- covariance matrix
- kernel machines
- sparse linear systems
- reproducing kernel hilbert space
- constrained optimization
- interior point methods
- geometric structure
- learning theory
- support vector
- symmetric positive definite
- genetic algorithm
- diffusion tensor
- neural network
- principal component analysis
- dynamic programming