Line search algorithms with guaranteed sufficient decrease
Jorge J. Morè, David J. Thuente · ACM Transactions on Mathematical Software · 1994
The development of software for minimization problems is often based on a line search method. We consider line search methods that satisfy sufficient decrease and curvature conditions, and formulate the problem of determining a point that satisfies these two conditions in terms of finding a point in a set T(μ) . We describe a search algorithm for this problem that produces a sequence of iterates that converge to a point in T(μ) and that, except for pathological cases, terminates in a finite number of steps. Numerical results for an implementation of the search algorithm on a set of test functions show that the algorithm terminates within a small number of iterations.
Cited by 528 works. doi:10.1145/192115.192132 Open-access PDF
Authors
References
- Numerical Methods for Unconstrained Optimization and Nonlinear Equations
- On the limited memory BFGS method for large scale optimization
- Global Convergence Properties of Conjugate Gradient Methods for Optimization
- Remark on “Algorithm 500: Minimization of Unconstrained Multivariate Functions [E4]”
- Descent Property and Global Convergence of the Fletcher—Reeves Method with Inexact Line Search
- Algorithm 500: Minimization of Unconstrained Multivariate Functions [E4]
- Global Convergence of a Cass of Quasi-Newton Methods on Convex Problems
- A view of line-searches
- Newton's method
- An efficient line search for nonlinear least squares