No label defined (Q3783086)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 4045496
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 4045496 |
Statements
1986
0 references
penalty method
0 references
variational inequalities
0 references
0.9121761918067932
0 references
0.8774147629737854
0 references
0.8686457276344299
0 references
0.8550357818603516
0 references
0.8385841846466064
0 references
The purpose of this paper is to propose a new penalty method for solving a class of variational inequalities of the following form: \[ (1)\quad Find\quad \bar x\in D,\quad \bar y\in F(\bar x)\quad such\quad that\quad <\bar y,x-\bar x>+\phi(x)-\phi (\bar x)\geq 0\quad \forall x\in D, \] where D is a closed convex subset of \(R^ n\), F is a maximal monotone (multivalued) mapping of \(R^ n\) into itself and \(\phi\) is a lower semicontinuous function on \(R^ n\). We use a penalty function P which is convex on \(R^ n\) and may be negative at points of the feasible region. This function thus, different from all other penalty functions, can measure the level of feasibility of each point. The value of this function at a point is a quantity of award or penalty for this point depending on its being feasible or not. The starting point may be arbitrary and a sequence of trial solutions will be determined so that they are feasible if the penalty function is not identical with zero on the feasible region. Otherwise the method becomes an exterior algorithm. In many cases of interest, a variational inequality for a monotone mapping (not necessary strictly) can be converted into a sequence of subproblems in each of which one calculates a solution of an unconstrained variational inequality for a certain strictly monotone mapping. This is very essential, since there are some available algorithms that may be used only in the problems with strictly monotone mappings. When \(F\equiv 0\) the problem (1) becomes one of minimizing the function \(\phi\) over the closed convex set D. Our method then allows to write it in the form of variational inequalities and this way is quite different from the usual techniques as it does not require convexity of the objective function.
0 references