Quadratic one-step smoothing Newton method for \(P_{0}\)-LCP without strict complementarity. (Q1406258)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Quadratic one-step smoothing Newton method for \(P_{0}\)-LCP without strict complementarity. |
scientific article |
Statements
Quadratic one-step smoothing Newton method for \(P_{0}\)-LCP without strict complementarity. (English)
0 references
9 September 2003
0 references
The linear complementarity problem with a matrix having all principal minors nonnegative is equivalent to a system of equations with Kanzow's smoothing function. This is solved by Newton's method involving only one linear system and only one Armijo-type line search. The algorithm has local quadratic convergence under nonsingularity condition. Boundedness of iterations is guaranteed by another condition. A comparison with methods from the literature is given and it is claimed that the algorithm has stronger convergence properties under weaker assumptions than many others.
0 references
linear complementarity problem
0 references
Newton's method
0 references
global convergence
0 references
quadratic convergence
0 references
smoothing
0 references
line search
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
0 references
