Cell-like maps. The problem of raising dimension

From MaRDI portal





\textit{R. C. Lacher} [Bull. Am. Math. Soc. 83, 495-552 (1977; Zbl 0364.54009)] gave an account of work up to 1977 involving cell-like mappings. Since then, many results have been obtained and the significance of cell-like mappings became even stronger than before. This paper brings an interesting survey of some recent developments in geometric topology related to cell-like mappings together with some results which probably appeared here for the first time. The titles of the sections provide quick information as to subjects treated in the paper: (1) Survey of the subject; (2) Approximative invertibility and the theorem on 3-dimensional manifolds; (3) Taylor's example and its consequences; (4) Theorem of R. D. Edwards on resolution of compacta; (5) Binary diameters and homological dimension; (6) Existence of mappings of spheres that are one-to-one over the image of a dense subset; (7) On \({\mathbb{Z}}_ p\)-homological dimension. The authors have written this paper as an introduction to the theory of homological dimension and geometric topology. The first section is a comprehensive introduction and historical account. In the second, the property that approximatively invertible mappings do not raise dimension is proved and used to prove that cell-like mappings of \(S^ 3\) do not raise the dimension, a result previously proved by \textit{G. Kozlowski} and \textit{J. J. Walsh} [Topology 22, 147-151 (1983; Zbl 0515.57008)]. The third section is devoted to a Taylor's example [\textit{J. L. Taylor}, Bull. Am. Math. Soc. 81, 629-632 (1975; Zbl 0316.55010)] with a variety of its consequences together with the following result: If M is a compact manifold, then the space cell M (of cellular subsets of M) has an open basis of contractible sets. The fourth section brings the proof of Edward's famous theorem on resolution of compacta: Every compactum of homological dimension n is the cell-like image of an n-dimensional compactum. The last three sections bring some recent results of the authors. The fifth section is devoted to the relation between homological dimension and binary diameter. Let \(f: X\to Y\) be a mapping of a compactum X. Then the n-dimensional binary diameter of f is defined as \(a_ nf=\inf_{\phi}dist(f,\phi)\), where \(\phi\) : \(X\to Y\) is any mapping such that dim Im \(\phi\leq n\). The (n,k)-dimensional binary diameter of f is defined as \(a^ n_ kf=\sup_{\phi}a_ k(f\phi)\), where \(\phi\) : X'\(\to X\) is a map of a compactum X'\(\subset X\) such that dim X'\(\leq n\). A sample result is: If X is embedded in a normed linear space then the following are equivalent: (a) Homological dimension of X is \(\leq n\); (b) \(a_ n^{n+1}(f)=0\) for any mapping \(f: X\to P\) into a finite polyhedron P. The sixth section is related to the Edwards-Henderson reduction of the CE problem to a certain unstable homotopy theory problem. For any \(f: X\to Y\) the set \(s(f)=\{y\in Y|\) \(f^{-1}(y)= \sin gleton\}\) is called the set of one-to-one points. The following theorem is proved: If \(m\geq n\geq 3\), then for every \(\alpha \in \pi_ m(S^ n)\) the following are equivalent: (1) \(\alpha\) is the suspension of a unique element of \(\pi_{m-1}(S^{n-1})\); (2) \(\alpha\) is represented by a mapping with three one-to-one points; (3) \(\alpha\) is represented by a mapping f such that s(f) is everywhere dense in \(S^ n\); (4) \(\alpha\) is represented by an irreducible mapping. The last section treats \(\dim_ p=\) the homological dimension with coefficients \({\mathbb{Z}}_ p\), and brings the following (quite unexpected) theorems: (1) For every prime p and for every integer \(n>2\) there exists a compactum X such that dim X\(=n\) while \(\dim_ p X=2\); (2) There exists an infinite-dimensional compactum X such that \(\dim_ p X=2\) for every prime p.











This page was built for publication: Cell-like maps. The problem of raising dimension

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3763269)