Globally convergent Polak-Ribière-Polyak conjugate gradient methods under a modified Wolfe line search.
Gaohang YuLutai GuanZengxin WeiPublished in: Appl. Math. Comput. (2009)
Keyphrases
- line search
- conjugate gradient
- globally convergent
- training algorithm
- convergence rate
- maximum likelihood estimation
- quadratic programming
- step size
- levenberg marquardt
- objective function
- faster convergence
- risk minimization
- back propagation
- linear program
- global convergence
- primal dual
- convergence speed
- maximum likelihood
- optimization methods
- linear programming
- evolutionary algorithm