Sets of large values of polynomial multi-correlation functions

V. Bergelson and R. Zelada · 2026

Read Sets of large values of polynomial multi-correlation functions by V. Bergelson and R. Zelada (2026) 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-2605-23050.txt.

Introduction

Bergelson and Zelada study sets of large returns for polynomial multi-correlation functions, defined by near maximal overlap of a measurable set with its shifts under commuting transformations. They show every such set is syndetic exactly when the polynomials are linearly independent, answering a question of Frantzikinakis and Kuca. Linear independence further implies the stronger almost IP star property. Combinatorially, for positive density sets, return sets are almost IP star, and when dimension exceeds polynomial count by more than one, this result is sharp.

Catalog record

CREATORV. Bergelson and R. Zelada
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE130152
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.23050

Browse the full library · About the Tower of Babel

Full text

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

Sets of large values of polynomial multi-correlation functions V. Bergelson and R. Zelada Abstract Let p 1 , … , p ℓ ∈ ℤ ⁡ [ x 1 , … , x d ] p_{1},...,p_{\ell}\in\mathbb{Z}[x_{1},...,x_{d}] be non-constant polynomials with zero constant term. The ergodic theoretical proofs of the polynomial [ 11 ] and the IP-polynomial [ 15 ] Szemerédi theorems as well as some of the ergodic-theoretical and combinatorial consequences of the Density Polynomial Hales-Jewett conjecture [ 19 ] naturally lead to the study of sets of large returns which are defined as R ϵ p 1 , … , p ℓ ​ ( A ) := { n → ∈ ℤ d | μ ⁡ ( A ∩ T 1 − p 1 ​ ( n → ) ​ A ∩ ⋯ ∩ T ℓ − p ℓ ​ ( n → ) ​ A ) > μ ℓ + 1 ​ ( A ) − ϵ } , R_{\epsilon}^{p_{1},...,p_{\ell}}(A):=\{\vec{n}\in\mathbb{Z}^{d}\,|\,\mu(A\cap T_{1}^{-p_{1}(\vec{n})}A\cap\cdots\cap T_{\ell}^{-p_{\ell}(\vec{n})}A)>\mu^{\ell+1}(A)-\epsilon\}, where T 1 , … , T ℓ T_{1},...,T_{\ell} are commuting and invertible measure preserving transformations on a probability space ( X , 𝒜 , μ ) (X,\mathcal{A},\mu) , A ∈ 𝒜 A\in\mathcal{A} , and ϵ > 0 \epsilon>0 . We obtain new results dealing with the fine properties of the sets of the form R ϵ p 1 , … , p ℓ ​ ( A ) R_{\epsilon}^{p_{1},...,p_{\ell}}(A) which in turn provide new combinatorial applications. Among other things, we show that every set of the form R ϵ p 1 , … , p ℓ ​ ( A ) R_{\epsilon}^{p_{1},...,p_{\ell}}(A) is syndetic if and only if the polynomials p 1 , … , p ℓ p_{1},...,p_{\ell} are linearly independent, answering a question asked by N. Frantzikinakis and B. Kuca [ 28 ] . Moreover, we prove that the linear independence of p 1 , … , p ℓ p_{1},...,p_{\ell} implies that every set of the form R ϵ p 1 , … , p ℓ ​ ( A ) R_{\epsilon}^{p_{1},...,p_{\ell}}(A) has the A-IP ∗ property (="almost" IP ∗ property), which is quite a bit stronger than syndeticity. The following is one of the new combinatorial results obtained in this paper. Suppose that p 1 , … , p ℓ p_{1},...,p_{\ell} are linearly independent. For any set E ⊆ ℤ D E\subseteq\mathbb{Z}^{D} with upper Banach density d ∗ ​ ( E ) > 0 d^{*}(E)>0 , any non-zero 𝐯 𝟏 , … , 𝐯 ℓ ∈ ℤ D {\bf v_{1},...,v_{\ell}}\in\mathbb{Z}^{D} , and any ϵ > 0 \epsilon>0 , the set S ϵ p 1 , … , p ℓ ​ ( E ) := { n → ∈ ℤ d | d ∗ ​ ( E ∩ ( E − p 1 ​ ( n → ) ​ 𝐯 𝟏 ) ∩ ⋯ ∩ ( E − p ℓ ​ ( n → ) ​ 𝐯 ℓ ) ) > ( d ∗ ​ ( E ) ) ℓ + 1 − ϵ } S_{\epsilon}^{p_{1},...,p_{\ell}}(E):=\{\vec{n}\in\mathbb{Z}^{d}\,|\,d^{*}(E\cap(E-p_{1}(\vec{n}){\bf v_{1}})\cap\cdots\cap(E-p_{\ell}(\vec{n}){\bf v_{\ell}}))>(d^{*}(E))^{\ell+1}-\epsilon\} is A-IP ∗ . Furthermore, we prove that when D > ℓ > 1 D>\ell>1 , this result is sharp in the sense that the A-IP ∗ property cannot be upgraded to IP ∗ . The techniques developed in this paper lead to some additional applications. For example, we show that an amplified form of the IP-polynomial Szemerédi theorem conjectured in [ 15 ] follows from the Density Polynomial Hales-Jewett conjecture. 1 Introduction 1.1 New and old results dealing with single and multiple recurrence for measure preserving \mathbb{Z} -actions We start our discussion with the formulation of the classical Khintchine recurrence theorem. Theorem 1.1 (Khintchine’s recurrence theorem [ 35 ] ) . For any invertible probability measure preserving system (X,\mathcal{A},\mu,T) , any A\in\mathcal{A} , and any \epsilon>0 , the set R_{\epsilon}(A,T):=\{n\in\mathbb{Z}\,|\,\mu(A\cap T^{-n}A)\geq\mu^{2}(A)-\epsilon\} (1.1) is syndetic, meaning that it has bounded gaps. In Khintchine’s original paper [ 35 ] , 1.1 was derived from the following variant of the ergodic theorem: Theorem 1.2 . Let (X,\mathcal{A},\mu,T) be an invertible probability measure preserving system and let A\in\mathcal{A} . Then, \lim_{N-M\rightarrow\infty}\frac{1}{N-M}\sum_{n=M+1}^{N}\mu(A\cap T^{-n}A)\geq\mu^{2}(A). (1.2) There is, however, an alternative approach to the proof of Khintchine’s recurrence theorem which provides additional information about the sets of the form R_{\epsilon}(A,T) . Let’s call E\subseteq\mathbb{Z} a \Delta^{*} set if for every strictly monotone sequence (n_{k})_{k\in\mathbb{N}} in \mathbb{Z} , the set \{n_{j}-n_{i}\,|\,j>i\} has a non-trivial intersection with E . It is not hard to see that any \Delta^{*} set is syndetic. On the other hand, not every syndetic set is \Delta^{*} (consider for example the set E=2\mathbb{Z}+1 ). So, the following proposition forms a non-trivial amplification of 1.1 . (This result is an easy corollary from the Enhanced Principle P (EPP) formulated and proved on page 34 of [ 21 ] .) Theorem 1.3 . Let (X,\mathcal{A},\mu,T) be an invertible probability measure preserving system, let A\in\mathcal{A} , and let \epsilon>0 . The set R_{\epsilon}(A,T)=\{n\in\mathbb{Z}\,|\,\mu(A\cap T^{-n}A)\geq\mu^{2}(A)-\epsilon\} is \Delta^{*} . Moreover, there exists an r:=r(\mu(A),\epsilon)\in\mathbb{N} with the property that for any distinct n_{1},...,n_{r}\in\mathbb{Z} , \{n_{j}-n_{i}\,|\,i<j,\,i,j\in\{1,...,r\}\}\cap R_{\epsilon}(A,T)\neq\emptyset. (1.3) It is worth mentioning that, by invoking Ramsey’s theorem, one can show that \Delta^{*} sets have the finite intersection property [ 31 , p. 179] , meaning that if E,F\subseteq\mathbb{Z} are \Delta^{*} sets, then E\cap F also is. It follows that if (X,\mathcal{A},\mu,T) and (Y,\mathcal{B},\nu,S) are invertible probability preserving systems, one has that for any A\in\mathcal{A} , B\in\mathcal{B} with \mu(A),\nu(B)>0 , and \epsilon^{\prime},\epsilon^{\prime\prime}>0 , the set R_{\epsilon^{\prime}}(A,T)\cap R_{\epsilon^{\prime\prime}}(B,S) (1.4) is \Delta^{*} (and, in particular, syndetic). Next we formulate the Furstenberg-Sárközy theorem [ 31 , Theorem 3.16] , [ 42 ] , which can be viewed as a polynomial extension of the classical Poincaré recurrence theorem. Theorem 1.4 (Furstenberg-Sárközy Theorem) . Let p\in\mathbb{Z}[x] be a non-constant polynomial with zero constant term and let (X,\mathcal{A},\mu,T) be an invertible probability preserving system. For any A\in\mathcal{A} with \mu(A)>0 , there are infinitely many n\in\mathbb{Z} with \mu(A\cap T^{-p(n)}A)>0 . A careful analysis of Furstenberg’s proof of 1.4 reveals that the following stronger result also holds. Theorem 1.5 . Let p\in\mathbb{Z}[x] be a non-constant polynomial with zero constant term and let (X,\mathcal{A},\mu,T) be an invertible probability preserving system. For any A\in\mathcal{A} with \mu(A)>0 and any \epsilon>0 , there is a natural number b such that \lim_{N-M\rightarrow\infty}\frac{1}{N-M}\sum_{n=M+1}^{N}\mu(A\cap T^{-p(bn)}A)\geq\mu^{2}(A)-\epsilon. (1.5) Corollary 1.6 . Let p\in\mathbb{Z}[x] be a non-constant polynomial with zero constant term and let (X,\mathcal{A},\mu,T) be an invertible probability preserving system. For any A\in\mathcal{A} and any \epsilon>0 , the set R_{\epsilon}^{p}(A,T):=\{n\in\mathbb{Z}\,|\,\mu(A\cap T^{-p(n)}A)>\mu^{2}(A)-\epsilon\} (1.6) is syndetic. In light of 1.5 it is natural to inquire if, in analogy with 1.3 , the sets of the form ( 1.6 ) are \Delta^{*} . It turns out that R_{\epsilon}^{p}(A,T) is not necessarily a \Delta^{*} set. For example, one can show that for any irrational number \alpha\in\mathbb{R} , the Lebesgue preserving transformation T_{\alpha}:[0,1)\rightarrow[0,1) defined by T_{\alpha}x=x+\alpha\mod 1 has the property that if A is an interval of length strictly smaller than 1/2 and p\in\mathbb{Z}[x] satisfies \deg(p)>1 , then the set R^{p}_{\epsilon}(A,T_{\alpha}) is not \Delta^{*} (for the case p(x)=x^{2} , see [ 31 , p. 177] ; see also [ 17 , Remark 5.4] ). On the other hand, one can show that sets of the form R_{\epsilon}^{p}(A,T) possess the IP ∗ property which is

Read the complete text