On dual surjunctivity and applications (Q2102724)

From MaRDI portal
scientific article
Language Label Description Also known as
English
On dual surjunctivity and applications
scientific article

    Statements

    On dual surjunctivity and applications (English)
    0 references
    0 references
    0 references
    0 references
    29 November 2022
    0 references
    For a finite alphabet \(A\) and a group \(G\) the shift action of \(G\) on \(A^G\) is a topological Bernoulli shift. If any injective, continuous (that is, depending on only finitely many coordinates), and \(G\)-equivariant map \(A^G\to A^G\) is automatically surjective, then \(G\) is called surjunctive. This notion was introduced by \textit{W. H. Gottschalk} [Lect. Notes Math. 318, 120--125 (1973; Zbl 0255.54035)], who raised the question of whether all groups are surjunctive. \textit{M. Gromov} [J. Eur. Math. Soc. (JEMS) 1, No. 2, 109--197 (1999; Zbl 0998.14001)] showed that the groups now called sofic are surjunctive. The sofic groups certainly form a large class, to the extent that no example of a non-sofic group is known. Here a reverse of surjunctivity is discussed. It is too much to simply swap surjective and injective around in the property, as there are counterexamples for \(G=\mathbb{Z}\). The weaker notion of pre-injectivity does however produce a meaningful question: For which groups \(G\) is a surjective, continuous, \(G\)-equivariant map pre-injective? \textit{E. F. Moore} [Proc. Symp. Appl. Math. 14, 17--33 (1962; Zbl 0126.32408)] and \textit{J. Myhill} [Proc. Am. Math. Soc. 14, 685--686 (1963; Zbl 0126.32501)] showed this for \(G=\mathbb{Z}^d\), and this was eventually extended to all amenable groups by \textit{T. G. Ceccherini-Silberstein} et al. [Ann. Inst. Fourier 49, No. 2, 673--685 (1999; Zbl 0920.43001)]. Indeed \textit{L. Bartholdi} [J. Eur. Math. Soc. (JEMS) 21, No. 10, 3191--3197 (2019; Zbl 1458.37017)] showed that this characterizes amenable groups. \textit{S. Capobianco} et al. [Lect. Notes Comput. Sci. 9664, 77--89 (2016; Zbl 1369.68261)] found a dual notion to that of \textit{W. H. Gottschalk} [loc. cit.], introducing a property strictly stronger than surjectivity called post-surjectivity, and showed that all sofic groups are dual surjunctive in that post-surjectivity implies pre-injectivity. This gives a strong version of the so-called ``Garden of Eden'' theorem, showing that a continuous \(G\)-equivariant map is injective if and only if it is post-surjective. All these matters are discussed here, with some of the arguments being simplified, some quantitative investigations of the properties explored, and the notion of post-surjectivity being studied for more general expansive dynamical systems including more general shifts.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    Gottschalk's conjecture
    0 references
    surjunctive groups
    0 references
    sofic groups
    0 references
    Kaplansky's direct finiteness
    0 references
    cellular automata
    0 references
    expansive algebraic actions
    0 references