A note on logarithmic mean equicontinuity

Dominik Kwietniak, Jian Li, Habibeh Pourmand · 2025

Read A note on logarithmic mean equicontinuity by Dominik Kwietniak, Jian Li, Habibeh Pourmand (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-04610.txt.

Introduction

Dominik Kwietniak and co-authors swap averages in the study of dynamical systems in a 2025 note. Ergodic theory watches where orbit averages of a system settle over time. Cesàro averages, the plain running mean, are the classic tool for that watch.

The note replaces them with logarithmic, harmonic-style averages along orbits. Unique ergodicity gains a fresh characterization through those gentler means. A new notion, logarithmic mean equicontinuity, is then defined for whole systems. It turns out to match ordinary mean equicontinuity exactly. Two averaging styles thus see the same calm systems after all.

Catalog record

CREATORDominik Kwietniak, Jian Li, Habibeh Pourmand
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE40 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-04610

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A note on logarithmic mean equicontinuity Dominik Kwietniak, Jian Li, Habibeh Pourmand Abstract . We study the set of harmonic limits of empirical measures in topological dynamical systems. We obtain a characterization of unique ergodicity based on logarithmic (harmonic) mean convergence in place of Cesáro convergence. We introduce logarithmic mean equicontinuity and show that a topological dynamical system is logarithmically mean equicontinuous if and only if it is mean equicontinuous. 2020 Mathematics Subject Classification 37B05, 37B25, 37A05 1. Introduction By a topological dynamical system, we mean a pair (X,T) where (X,d) is a compact metric space and T\colon X\to X is a continuous map. We say that a topological dynamical system (X,T) is equicontinuous if for every \varepsilon>0 there exists \delta>0 such that for any x,y\in X with d(x,y)<\delta , one has d(T^{n}x,T^{n}y)<\varepsilon for all n\geq 0 . Equicontinuity ensures stable, predictable behaviour as nearby points remain close under all iterations. Equicontinuous systems play a fundamental role in topological dynamics, and every surjective equicontinuous system is conjugate to an isometric system given by a rotation on a compact topological group. When studying dynamical systems with discrete spectrum, Fomin [ 5 ] introduced a weak form of equicontinuity, called mean-L-stability . A topological dynamical system (X,T) is mean-L-stable if for every \varepsilon>0 there exists \delta>0 such that for any x,y\in X with d(x,y)<\delta and for any n\geq 0 outside a set of upper density less than \varepsilon , one has d(T^{n}x,T^{n}y)<\varepsilon . It had been an open problem whether any ergodic invariant measure on a mean-L-stable system has discrete spectrum, see [ 16 ] . In [ 12 ] , Li et al. introduced the concept of mean equicontinuity. A topological dynamical system (X,T) is mean equicontinuous if for every \varepsilon>0 there exists \delta>0 such that for any x,y\in X with d(x,y)<\delta , one has \limsup_{n\to\infty}\frac{1}{n}\sum_{k=1}^{n}d(T^{k-1}x,T^{k-1}y)<\varepsilon. The authors of [ 12 ] also showed that mean equicontinuity is equivalent to mean-L-stability and confirmed that every ergodic invariant measure of a mean-L-stable system has discrete spectrum. In [ 7 ] , García-Ramos, introduced the concept of mean equicontinuity for an invariant measure of a topological dynamical system. He showed that an ergodic invariant measure is mean equicontinuous if and only if it has discrete spectrum. Recently in [ 9 ] , Huang et al. showed that any (not necessarily ergodic) invariant measure is mean equicontinuous if and only if it has discrete spectrum. A direct consequence of this result is that any invariant measure on a mean-L-stable system has discrete spectrum. In 2010, Sarnak [ 15 ] formulated the following conjecture: for every topological dynamical system (X,T) with zero topological entropy, every x\in X , and every continuous function f\colon X\to\mathbb{R} , \lim_{n\to\infty}\frac{1}{n}\sum_{k=1}^{n}f(T^{k}x){\bm{\mu}}(n)=0, where {\bm{\mu}}(n) is the classical Möbius function. We refer the reader to the survey [ 4 ] and references therein for the recent developments concerning this conjecture. In [ 17 ] , Tao studied the following logarithmic version of Sarnak’s conjecture: for every topological dynamical system (X,T) with zero topological entropy, every x\in X , and every continuous function f\colon X\to\mathbb{R} , one has \lim_{n\to\infty}\frac{1}{\log n}\sum_{k=1}^{n}\frac{1}{k}f(T^{k}x){\bm{\mu}}(n)=0. Tao proved that the logarithmic versions of Sarnak’s and Chowla’s conjectures are equivalent. In [ 8 ] Gomilko et al. showed that Sarnak’s conjecture implies the Chowla’s conjecture along a subsequence. Frantzikinakis and Host [ 6 ] verified the logarithmic version of Sarnak’s conjecture for a large class of topological dynamical systems, in particular for all uniquely ergodic systems with zero entropy. In view of these results, it is natural to consider the logarithmic version of mean equicontinuity. We say that a topological dynamical system is logarithmically mean equicontinuous if for any \varepsilon>0 there exists \delta>0 such that for any x,y\in X with d(x,y)<\delta , one has \limsup_{n\to\infty}\frac{1}{H_{n}}\sum_{k=1}^{n}\frac{1}{k}d(T^{k-1}x,T^{k-1}y)<\varepsilon, where H_{n}=\sum_{k=1}^{n}\frac{1}{k} . It is easy to see that mean equicontinuity implies logarithmic mean equicontinuity. The main result of this note is to show that the converse is also true, and hence logarithmic mean equicontinuity is in fact equivalent to mean equicontinuity. Theorem 1.1 . A topological dynamical system (X,T) is logarithmically mean equicontinuous if and only if it is mean equicontinuous. The proof uses properties of the harmonic limits of empirical measures studied in [ 6 ] , [ 8 ] , and [ 14 ] . In order to get our proof we need to study limits of empirical measures with logarithmic (harmonic) mean convergence replacing Cesàro convergence. That is, we study convergence of subsequences of logarithmically averaged empirical measures and compare them with arithmetically averaged empirical measures. In particular, we obtain a characterization of unique ergodicity by logarithmic mean convergence, see Theorem 2.15 . We also consider the logarithmic version of mean equicontinuity for invariant measures, and show that an ergodic measure is logarithmically mean equicontinuous if and only if it is mean equicontinuous, see Theorem 4.6 . Note that in general, the results of logarithmic averaging and Cesàro averaging of empirical measures are different (see [ 8 ] ). 2. Invariant measures and unique ergodicity In this section, we first recall results concerning the classical empirical and invariant measures of topological dynamical systems and a characterization of unique ergodicity. Then we show that similar results hold for harmonic limits of empirical measures and we obtain a characterization of unique ergodicity by logarithmic mean convergence. We think that this approach is of independent interest. Let (X,d) be a compact metric space and C(X,\mathbb{R}) be the space of continuous maps from X to \mathbb{R} with the supremum norm \|\cdot\|_{\infty} . Let 2^{X} be the hyperspace of X , that is, 2^{X} is the collection of non-empty closed subsets of X . The Hausdorff metric on 2^{X} is defined by d_{H}(A,B)=\inf\{\varepsilon>0\colon A\subset\{y\in X:d(y,B)<\varepsilon\}\ \&\ B\subset\{x\in X:d(x,A)<\varepsilon\}\}. Here d(z,C) for z\in X and C\in 2^{X} means \inf\{d(z,c):c\in C\} . Then (2^{X},d_{H}) is a compact metric space. Let \mathcal{B}(X) be the collection of all Borel subsets of X and M(X) be the collection of all probability measures on (X,\mathcal{B}(X)) . The weak ∗ -topology on M(X) is the smallest topology making each of the maps \mu\mapsto\int f\mathop{}\!\mathrm{d}\mu ( f\in C(X,\mathbb{R}) ) continuous. Then M(X) is a compact metrizable space. To define a metric on M(X) , one fixes a countable collection \{f_{j}\colon j\geq 1\} in C(X,\mathbb{R}) such that the linear span of \{f_{j}\colon j\geq 1\} is dense in C(X,\mathbb{R}) and for each j\geq 1 we have \|f\|_{\infty}\leq 1 and |f_{j}(x)-f_{j}(y)|\leq d(x,y) for all x,y\in X , see e.g. [ 2 , Theorem 11.2.4] . Then for \mu,\nu\in M(X) a compatible metric on M(X) is given by \rho(\mu,\nu)=\sum_{j\geq 1}\frac{1}{2^{j}}\biggl|\int f_{j}\mathop{}\!\mathrm{d}\mu-\int f_{j}\mathop{}\!\mathrm{d}\nu\biggr|. Let (X,T) be a topological dynamical system. We say that \mu\in M(X) is T -invariant if \mu(B)=\mu(T^{-1}B) for every B\in\mathcal{B}(X) . If a T -invariant measure \mu is such that for every B\in\mathcal{B}(X) with T^{-1}B=B we have either \mu(B)=0 or \mu(B)=1 , then we say that \mu is ergodic . Let M_{T}(X) and M^{\mathrm{erg}}_{T}(X) stand for the collection of T -invariant measures and ergodic measures, respectively. Given x\in X , we define the empirical measure along the orbit of x to be an arithmetic average of Dirac measures concentrated on the consecutive points along the orbit, that is, the empirical measure is the measure \frac{1}{n-m}\sum_{k=m+1}^{n}\delta_{T^{k-1}x}, where

Read the complete text