A bridge between U -frequent hypercyclicity and frequent hypercyclicity

For any sequence of positive real numbers \((a_n)_{n\in\mathbb{N}}\), we can define the lower weighted density \(\underline{d}_{a}\) and the upper weighted density \(\overline{d}_{a}\) on subsets \(I\subseteq\mathbb{N}\) by \[ \underline{d}_{a}(I)=\liminf_{N\to\infty}\frac{\sum_{n\in I\cap[0;N]}a_n}{\sum_{n=0}^{N}a_n}\text{ and }\overline{d}_{a}(I)=\limsup_{N\to\infty}\frac{\sum_{n\in I\cap[0;N]}a_n}{\sum_{n=0}^{N}a_n}. \] Note that the classical lower \(\underline{d}\) and upper \(\overline{d}\) densities are recovered with the constant sequence equal to \(1\). Moreover, if \(\mathcal{A}\) denotes the set of positive sequences decreasing to zero and so that \(\sum_{n\in\mathbb{N}}a_n=\infty\), then for any \((a_n)_{n\in\mathbb{N}}\in\mathcal{A}\) we have the following ordering for any \(I\subseteq\mathbb{N}\): \[ \underline{d}(I)\leq \underline{d}_{a}(I)\leq \overline{d}_{a}(I)\leq \overline{d}(I). \] The author is motivated by the study of dynamical properties of continuous linear operators \(T\) on infinite-dimensional Fréchet spaces \(X\). In particular, for \(x\in X\) one may want to quantify how irregular is the behavior of the orbit \(\mathcal{O}(x,T):=\{T^nx;n\in \mathbb{N}\}\) by evaluating the size of the return sets \(N(x,U):=\{n\in\mathbb{N}: T^nx\in U\}\) for every non-empty open set \(U\) of \(X\). Such studies have become very active since the article of \textit{F. Bayart} and \textit{S. Grivaux} [Trans. Am. Math. Soc. 358, No. 11, 5083--5117 (2006; Zbl 1115.47005)] who introduced the concept of frequent hypercyclicity. Later, Shkarin in 2009 [\textit{S. Shkarin}, Proc. Am. Math. Soc. 137, No. 1, 123--134 (2009; Zbl 1161.47007)] defined the weaker notion of \(\mathcal{U}\)-frequent hypercyclicity. Let us recall that an operator is called frequently hypercyclic (respectively, \(\mathcal{U}\)-frequently hypercyclic) if there exists a vector \(x\in X\) such that for every non-empty open subset \(U\) of \(X\), the classical lower (respectively, upper) density of the return set \(N(x,U)\) is positive, i.e., \(\underline{d}(N(x,U))>0\) (resp., \(\overline{d}(N(x,U))>0\)). Such a vector \(x\) is called a frequently hypercyclic vector (resp., \(\mathcal{U}\)-frequently hypercyclic vector) and the set of all such vectors is denoted by \(FHC(T)\) (resp., \(UFHC(T)\)). \textit{R. Ernst} and \textit{A. Mouze} [Ergodic Theory Dyn. Syst. 39, No. 4, 898--924 (2019; Zbl 1482.47010)] studied frequent hypercyclicity notions associated to weighted densities with weight \((a_n)_{n\in\mathcal{N}}\in\mathcal{A}\) defined above and we denote by \(FHC_a(T)\) (resp., \(UFHC_a(T)\)) the set of vectors such that every return set has positive lower (resp., upper) weighted density, i.e., \(\underline{d}_a(N(x,U))>0\) (resp., \(\overline{d}_a(N(x,U))>0\)) for every non-empty open subset \(U\subseteq X\). In fact, for weights in \(\mathcal{A}\), the inequality on the densities shows that all frequent hypercyclicity notions associated to weighted densities are intermediate notions between frequent and \(\mathcal{U}\)-frequent hypercyclicity. Moreover, it appeared that frequently hypercyclic operators and \(\mathcal{U}\)-frequent hypercyclicity have some important differences. The aim of the author is to use weighted densities to create a ``bridge between these two notions and to better understand their similarities and differences. The author begins with proving \[ UFHC(T)=\cup_{a\in\mathcal{A}}FHC_a(T)\text{ and }FHC(T)=\cap_{a\in\mathcal{A}}UFHC_a(T) \] which yields that every \(\mathcal{U}\)-frequently operator is automatically frequently hypercyclic for some \(a\in\mathcal{A}\). This result shows a link between the two notions of frequent and \(\mathcal{U}\)-frequent hypercyclicity. This last result makes us wonder if every \(\mathcal{U}\)-frequently hypercyclic operator for any \(a\in\mathcal{A}\) is automatically frequently hypercyclic. The author studies this question and concludes the article with a technically involved example of an operator on \(\ell^1\) that is \(\mathcal{U}\)-frequently hypercyclic for any \(a\in\mathcal{A}\) but not frequently hypercyclic. This operator is an operator of C-type, which was first introduced by \textit{Q. Menet} [Trans. Am. Math. Soc. 369, No. 7, 4977--4994 (2017; Zbl 1454.47014)] to prove that there exist chaotic operators that are not frequently hypercyclic.

Recommendations

Cites work

Cited in

(8)

This page was built for publication: A bridge between \(\mathcal{U} \)-frequent hypercyclicity and frequent hypercyclicity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2009283)