Persistence of zero sets

From MaRDI portal



Abstract: We study robust properties of zero sets of continuous maps f:XomathbbRn. Formally, we analyze the family Zr(f)=g−1(0):,,|g−f|<r of all zero sets of all continuous maps g closer to f than r in the max-norm. The fundamental geometric property of Zr(f) is that all its zero sets lie outside of A:=x:,|f(x)|ger. We claim that once the space A is fixed, Zr(f) is emph{fully} determined by an element of a so-called cohomotopy group which---by a recent result---is computable whenever the dimension of X is at most 2n−3. More explicitly, the element is a homotopy class of a map from A or X/A into a sphere. By considering all r>0 simultaneously, the pointed cohomotopy groups form a persistence module---a structure leading to the persistence diagrams as in the case of emph{persistent homology} or emph{well groups}. Eventually, we get a descriptor of persistent robust properties of zero sets that has better descriptive power (Theorem A) and better computability status (Theorem B) than the established well diagrams. Moreover, if we endow every point of each zero set with gradients of the perturbation, the robust description of the zero sets by elements of cohomotopy groups is in some sense the best possible (Theorem C).


Given a vector valued continuous map \(f:X \to \mathbb{R}^n\) it is natural to be interested in its zero set and in the topological properties of it. Often, however, we only have access to an \(r\) \textit{perturbation} \(g\) of \(f\) (\(r\) a positive real number), i.e. a function such that \(\| f-g \|_\infty <r\). The goal of this paper is the study of the set \(Z_{<r}(f)\) of the zero sets of such perturbations. The paper completes preceding work [\textit{P. Franek} and \textit{M. Krčál}, J. ACM 62, No. 4, Article No. 26, 19 p. (2015; \url{doi:10.1145/2751524})] on the algorithmic decidability of such sets. A related problem had been studied, with varying \(r\), through \textit{well modules} and \textit{zigzag persistence} in [\textit{H. Edelsbrunner} et al., Found. Comput. Math. 11, No. 3, 345--361 (2011; Zbl 1232.55012)]. Surprisingly enough, \(Z_{<r}\) is determined by the set \(A_r\) of points \(x\) on which \(f(x)\geq r\) and by the homotopy class of the normalizing function \(A \to \mathbb{S}^{n-1}\) (Theorem A). On the filtration given by the sets \(A_r\) the authors build a \textit{cohomotopy persistence module} of whose computation they prove polynomial complexity (Theorem B). The research then specializes to manifolds \(X\) and to maps \(f:X \to \mathbb{R}^n\) for which 0 is a regular value. Theorem C characterizes the set, analogous to \(Z_{<r}\) for \textit{regular} \(r\)-perturbations, in terms of cohomotopy (sub)groups when the dimension of the manifold is \(\leq 2n-3\). These are the main results, but the paper has a much wider horizon. All arguments are finely detailed and well commented.











This page was built for publication: Persistence of zero sets

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1689742)