A nonsmooth Robinson's inverse function theorem in Banach spaces (Q263194)

The paper is concerned with inverse mapping results for set-valued mappings \(F:X\rightrightarrows Y\), where \(X\) and \(Y\) are Banach spaces, in the spirit of those proved in the book by \textit{A. L. Dontchev} and \textit{R. T. Rockafellar} [Implicit functions and solution mappings. A view from variational analysis, 2nd updated ed. Springer Series in Operations Research and Financial Engineering. New York, NY: Springer (2014; Zbl 1337.26003)]. For \(\bar y\in F(\bar x)\) one says that \(\tilde F:X\rightrightarrows Y\) is a graphical localization at \(\bar x\) for \(\bar y\) if, for some neighborhoods \(U\) of \(\bar x\) and \(V\) of \(\bar y\),\; gph\(\tilde F=(U\times V)\cap\)gph\(F\), that is \(\{(x,y)\in X\times Y : y\in\tilde F(x)\}=\) \(\{(x,y)\in U\times V : y\in F(x)\}\). The mapping \(F\) is said to be strongly metrically regular at \(\bar x\) for \(\bar y\) if \(F^{-1}\) has a Lipschitz continuous single-valued graphical localization around \(\bar y\) for \(\bar x\), i.\,e. a graphical localization which is a Lipschitz continuous single-valued function in a neighborhood of \(\bar y\). The main result of the paper (Theorem 3) is concerned with a single-valued function \(f:X\to Y\), a set-valued mapping \(F:X\rightrightarrows Y\) and a point \((\bar x,\bar y)\in X\times Y\) such that \(\bar y\in f(\bar x)+F(\bar x)\). One supposes that there exist a convex subset \(\mathcal A\) of \(\mathcal L(X,Y)\) and \(c>0\) such that the following conditions are satisfied: (A)\; \; there exists \(r>0\) such that for every \(u,v\in \mathbb B_r(\bar x)\) (closed ball) there exists \(A\in\mathcal A\) such that \( \|f(v)-f(u)-A(v-u)\|\leq c\|v-u\|\); (B)\;\; for every \(A\in \mathcal A\) the mapping \(G_A(x):= f(\bar x)+A(x-\bar x)+F(x),\, x\in X\), is strongly metrically regular at \(\bar x\) for \(\bar y\) and, moreover, any single-valued graphical localization \(s_A\) of \(G_A^{-1}\) around \(\bar y\) for \(\bar x\) satisfies \((c+\chi(\mathcal A))\sup_{A\in\mathcal A}\)lip\((s_A,\bar y)<1\). Then, the mapping \(f + F\) is strongly metrically regular at \(\bar x\) for \(\bar y\). Here \(\chi(\mathcal A)\) is the Kuratowski measure of noncompactness of the set \(\mathcal A\) and lip\((s_A,\bar y)\) is the Lipschitz modulus of \(s_A\) at \(\bar y\), i.\,e. the infimum of all \(l>0\) such that \(s_A\) is Lipschitz with constant \(l\) on some neighborhood of \(\bar y\). This very general result contains as particular cases some inverse mapping theorems proved by \textit{A. D. Ioffe}, [Trans. Am. Math. Soc. 266, 1--56 (1981; Zbl 0651.58007)], \textit{A. F. Izmailov}, [Math. Program. 147, No. 1--2 (A), 581--590 (2014; Zbl 1301.49043)], and \textit{Z. Páles}, [J. Math. Anal. Appl. 209, No. 1, 202--220 (1997; Zbl 0880.58002)]. As the authors warn: ``The authors are aware of the fact that assumptions (A) and (B) in Theorem 3 above may be difficult to check in the general case considered. However, we expect that in special cases, e.g., for optimal control problems, these assumptions can be reduced to numerically tractable conditions. This is a subject of our continuing research on this topic.''