A new subspace minimization conjugate gradient method with nonmonotone line search for unconstrained optimization.
Ming LiHongwei LiuZexian LiuPublished in: Numer. Algorithms (2018)
Keyphrases
- line search
- conjugate gradient
- trust region
- objective function
- training algorithm
- convergence rate
- maximum likelihood estimation
- levenberg marquardt
- faster convergence
- dimensionality reduction
- step size
- low dimensional
- variational inequalities
- linear program
- constrained optimization
- quadratic programming
- feature space
- high dimensional data
- global optimum
- high dimensional
- loss function
- principal component analysis
- global convergence
- support vector
- kernel machines
- genetic algorithm
- reproducing kernel hilbert space
- convergence speed
- optimization algorithm
- multi view
- data points
- artificial neural networks