scientific article; zbMATH DE number 35090

From MaRDI portal
Publication:3989706





The author proves a ``multivalued version of a result of F. E. Browder: Let \(K\) be a closed, convex subset of a Hilbert space \(H\), \(x_ 0\in K\) and \(T\) be a nonexpansive mapping from \(K\) into the set \(\mathbb{K}\) of nonempty compact subsets of \(K\) (Hausdorff metric on \(\mathbb{K}\)). Assume that there is a \(z\in K\) with \(\text{Fix}(T)=\{z\}\) and \(Tz=\{z\}\), then \(\text{Fix}(\{\lambda x_ 0\}+(1-\lambda)T)\) converges in the sense of Mosco to \(\text{Fix}(T)\) as \(\lambda\to 1-\), i.e. we have for each sequence \((\lambda_ n)\in(0,1)^{\mathbf N}\): \[ \{x\in H:\;\exists(x_ n)\in K^{\mathbf N}\text{ with } x_ n\in\text{Fix}(\{\lambda_ n x_ 0\}+(1-\lambda_ n)T)\text{ and } x_ n\to x\}=\{z\}= \] \(\{x\in H\): \(\exists\) a subsequence \((\lambda_ j)\) of \((\lambda_ n)\) and \((x_ j)\in K^{\mathbf N}\) with \(x_ j\in\text{Fix}(\{\lambda_ j x_ 0\}+(1- \lambda_ j)T)\) and \(x_ j\to x\) (weakly)\(\}\).





Cited in
(33)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3989706)