Continuity concepts for set-valued functions and a fundamental duality formula for set-valued optimization (Q1759940)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Continuity concepts for set-valued functions and a fundamental duality formula for set-valued optimization |
scientific article |
Statements
Continuity concepts for set-valued functions and a fundamental duality formula for set-valued optimization (English)
0 references
22 November 2012
0 references
Let \(X,Z\) be topological vector spaces with duals \(X^*, Z^*\). Suppose that \(Z\) is ordered by a closed convex cone \(C\) and let \(\mathcal F(Z,C)=\{A\subset Z : A = cl(A+C)\}\). Define an order on \(\mathcal P(Z) \) by \(A\preccurlyeq_C B\iff B\subset A+C\). Then for \(A,B\in \mathcal F(Z,C)\), \(A\preccurlyeq_C B\iff B\subset A\), \(\mathcal F(Z,C)\) is a complete lattice with respect to \(\preccurlyeq_C\) and the ordered space \((Z,\leq_C)\) can be embedded into \(\mathcal F(Z,C)\) by the mapping \(z\mapsto \{z\}+C\). A mapping \(f:X\to\mathcal P(Z)\) is called convex if \(f(t x_1+(1-t)x_2)\preccurlyeq_C tf(x_1) + (1-t) f(x_2)\), \(\forall x_1,x_2\in X\), \(\forall t\in [0,1]\). If \(f\) takes values in \(\mathcal F(Z,C)\), then \(f\) is convex iff its graph is a convex subset of \(X\times Z\). A point \(x_0\in X\) is called an efficient point for \(f\) if there exists a neighborhood \(U\) of \(x_0\) and a bounded set \(B\subset Z\) such that \(\forall x\in U\), \(f(x)\cap B\neq\emptyset\). The authors consider various continuity notions for set-valued mappings (upper continuity, lower continuity, Hausdorff continuity), lattice semi-continuity properties for mappings taking values in a complete lattice, study the relations between these continuity notions as well as the continuity properties of a mapping \(f\) at an efficient point \(x_0\). Based on these properties and using the negative conjugate \(-f^*\) of a function \(f:X\to\mathcal F(Z,C)\) and its scalarization in a direction \(z^*\in Z^*\), they prove a very general duality result for set-valued optimization problems.
0 references
set-valued map
0 references
set-valued optimization
0 references
convex vector function
0 references
semi-continuity
0 references
Legendre-Fenchel conjugate
0 references
duality in optimization
0 references
ordered vector space
0 references
lattice
0 references