• Aucun résultat trouvé

Weighted U-frequent hypercyclicity

N/A
N/A
Protected

Academic year: 2021

Partager "Weighted U-frequent hypercyclicity"

Copied!
11
0
0

Texte intégral

(1)

Weighted

U

-frequent hypercyclicity

Céline ESSER [email protected]

University of Liège – Institute of Mathematics

Lens – October 24, 2018

Joint work with

(2)

Introduction and statement of the problem

LetT : X → X be a linear operator (X separable Banach space,dim X = ∞).

• TisU-frequently hypercyclic (ufhc)if∃x ∈ X such that∀U 6= ∅open

d {n ≥ 0 : Tnx ∈ U } > 0 .

The set of such points is denotedUFHC(T )(Shkarin 2009).

• Tisreiteratively hypercyclic (rhc)if∃x ∈ Xsuch that∀U 6= ∅open

Bd {n ≥ 0 : Tnx ∈ U } > 0 .

(3)

ForE ⊆ N, the densities used are

• the upper densityd(E) = lim sup n→+∞

# E ∩ [0, n]

n + 1 • the upper Banach densityBd(E) = lim

k→+∞lim supn→+∞

# E ∩ [n, n + k]

k + 1

One hasd(E) ≤ Bd(E)hence

T ufhc =⇒ T rhc

but one can prove thatTrhc ; Tufhc (Bès, Menet, Peris, Puig 2016)

(4)

Weighted upper densities

Leta = (an)n∈Nbe a sequence of positive real numbers such thatP

n∈Nan= +∞.

Thea-upper densityof a setE ⊆ Nis defined by

da(E) = lim sup

n→+∞ n X k=0 ak1E(k) n X k=0 ak

An operatorTisa-upper frequently hypercyclic (ufhca)if∃x ∈ X such that∀U 6= ∅

open,

da {n ≥ 0 : Tnx ∈ U } > 0 .

The set of such points is denotedUFHCa(T )and is a residual set

(5)

Remarks

• If an

bn & 0, thenda≤ db. Hence ufhca⇒ufhcb (Ernst, Mouze 2017)

• Ifan= 1for alln ∈ N, thenda= d. Hence ufhca⇔ufhc

• Letνn= an

Pn−1

k=0ak

. If there isC > 0such thatνn ≥ Cfor alln ∈ N, then

da(E) > 0as soon as#E = +∞

Consequences

• (H1) We consider sequencesa = (an)n∈Nsuch thatan % +∞ • (H2) We impose thatνn& 0

(6)

Example 1

For anyα > 0, the sequence(nα)n∈Nbelongs toA:

• nα% +∞ • νn= Pn−1nα k=1kα ∼ nα nα+1 & 0

Example 2

For anyα > 1, the sequence(αn)n∈Ndoes not belong toA:

νn= αn Pn−1 k=1αk = α n α−αn 1−α = (1 − α) 1 α1−n− 1 → α − 1

(7)

Comparison of the densities

Result

• For everya ∈ A, one hasd ≤ da.

• For everyα > 0, the sequencea = (nα)

n∈N∈ Asatisfies 1

1 + αda≤ d ≤ da.

Consequence

For every operatorT, there isa ∈ Asuch thatUFHC(T ) = UFHCa(T ). In particular,

UFHC(T ) = \

a∈A

UFHCa(T )

and

(8)

Result

• For everya ∈ A, one hasda≤ Bd.

• Let(Ek)k∈Nbe a sequence of subsets ofN. There existsa ∈ Asuch that

Bd(Ek) ≤ e da(Ek) ∀k ∈ N .

Consequence

For every operatorT,

RHC(T ) = [

a∈A

UFHCa(T )

and

(9)

IfT is rhc, thenRHC(T ) = HC(T )(Bès, Menet, Peris, Puig 2016)

Result

For everya ∈ A, one hasUFHCa(2B) 6= HC(2B).

However, ifT is rhc, then

HC(T ) = [

a∈A

UFHCa(T )

Chaos and ufhc

a

One has(Menet 2017) • chaos⇒rhc

• chaos;ufhc

Result

For everya ∈ A, there is a chaotic operator on`1which is not ufhc a.

(10)

Product of operators

Question: Given an operatorTwhich is ufhc / ufhca/ rhc, what can we say about the

operatorT × · · · × T?

Result

IfT is ufhca, thenT × · · · × T is ufhca.

Keys:(using ideas from Bayart, Rusza 2015&Bès, Menet, Peris, Puig 2016) • ufhca⇒weakly-mixing

• Ifda(E) > 0, then there isδ > 0such that the set

k ≥ 0 : da E ∩ (E − k)) > δ

is syndetic

Consequence

(11)

Main references

F. Bayart and I. Z. Ruzsa

Difference sets and frequently hypercyclic weighted shifts

Ergod. Theory Dyn. Syst.,35, 691–709, 2015.

J. Bès, Q. Menet, A. Peris and Y. Puig

Recurrence properties of hypercyclic operators

Math. Ann.366, 545–572, 2016.

A. Bonilla and K-G. Grosse-Erdmann

Upper frequent hypercyclicity and related notions

Rev. Mat. Complut.,31, 673–711, 2018.

R. Ernst and A. Mouze

A quantitative interpretation of the frequent hypercyclicity criterion

Ergod. Theory Dyn. Syst., https://doi.org/10.1017/etds.2017.55, 2017.

Q. Menet

Linear chaos and frequent hypercyclicity

Trans. Amer. Math. Soc.369, 4977–4994, 2017.

S. Shkarin

On the spectrum of frequently hypercyclic operators

Références

Documents relatifs

, On the zeros of functions with finite Dirichlet integral and some related function

Actually we show that an operator which satisfies the Frequent Hypercyclicity Criterion is necessarily A-frequently hypercyclic, where A refers to some weighted densities sharper

Corollary 2.27 tells us that this is not the case: there exists two multiples of the same weighted shift, which satisfy the classical Frequent Hypercyclicity Criterion, and so

The aim of the present paper is to try to fill the gap between frequently hypercyclic operators and these somewhat extreme examples, by investigating the frequency of hypercyclicity

Indeed, our methodology recovers classical local PDEs-based levelings in image processing, gener- alizes them to nonlocal configurations and extends them to process any discrete

Secondly, based on the same discrete derivatives, we formulate mathemat- ical morphology operators (dilation and erosion) which can be used to perform several morphological processes

In this paper we propose an adaptation of PDEs level sets over weighted graphs of arbitrary structure, based on PdEs and us- ing a framework of discrete operators.. A general PDEs

Regarding the other approaches of the literature, our approach to transpose PDEs on surfaces or point clouds (represented as graphs) has many advantages: no parametrization is