Algorithm 851
William W. Hager, Hongchao Zhang · ACM Transactions on Mathematical Software · 2006
Recently, a new nonlinear conjugate gradient scheme was developed which satisfies the descent condition g T k d k ≤ −7/8 ‖ g k ‖ 2 and which is globally convergent whenever the line search fulfills the Wolfe conditions. This article studies the convergence behavior of the algorithm; extensive numerical tests and comparisons with other methods for large-scale unconstrained optimization are given.
Cited by 410 works. doi:10.1145/1132973.1132979
Authors
References
- Methods of conjugate gradients for solving linear systems
- Practical Methods of Optimization
- Function minimization by conjugate gradients
- On the limited memory BFGS method for large scale optimization
- Restart procedures for the conjugate gradient method
- Updating Quasi-Newton Matrices with Limited Storage
- Benchmarking optimization software with performance profiles
- Applied numerical linear algebra
- Practical Methods of Optimization. Volume 1: Unconstrained Optimization
- Convergence Conditions for Ascent Methods