Another form of KKM type theorem and its applications on generalized convex spaces (Q2505270): Difference between revisions

The authors formulate a version of the KKM theorem for \(G\)-convex spaces [cf.\ \textit{S.~Park}, Korean J.\ Comp.\ Appl.\ Math.\ 7, 1--28 (2000; Zbl 0959.47035)]. The main result reads as follows. Theorem. Let \((X,D;\Gamma)\) be a \(G\)-convex space, \(Y\) a topological space, \(S:X\to Y\) and \(A:D\to Y\) maps, and \(K\) a nonempty compact subset of \(X\) such that \(S\) is upper semicontinuous on \(X\) and for each nonempty finite subset \(N\) of \(D\) there exists an \(L\) such that for every \(z\in L\setminus K\) there exists an \(x\in D\cap L\) satisfying \(S(z)\cap\overline{A(x)}=\emptyset\). Then either there exists an \(x_0\in K\) such that \(\overline{A(x)}\cap S(x_0)\not=\emptyset\) for all \(x\in D\), or there exists a finite set \(\{x_1,\dotsc,x_n\}\) in \(D\) and \(x\in\Gamma(x_1,\dotsc,x_n)\) such that \(S(x)\subset Y\setminus\bigcup_{i=1}^nA(x_i)\). This reviewer is somewhat skeptical as to whether this is really interesting.