On multiplicative recurrence along linear patterns
DIMITRIOS CHARAMARAS, ANDREAS MOUNTAKIS AND KONSTANTINOS TSINAS · 2024
Read On multiplicative recurrence along linear patterns by DIMITRIOS CHARAMARAS, ANDREAS MOUNTAKIS AND KONSTANTINOS TSINAS (2024) 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-2412-03504.txt.
Introduction
Dimitrios Charamaras and co-authors settle which linear patterns force a return in multiplying systems, in a 2024 math paper. Recurrence asks when a system's path must come back near its start. For systems driven by multiplication, past work found some ratio sets that count as return sets. A key test for such a set runs through special functions that twist each number by a fixed rule.
The authors pin down every set of four whole numbers that passes that test. Their result extends a known case for the simple pair of neighbors, and work by four other scholars. They then prove more: under the same conditions, the ratio set truly is a return set for a broad class of systems. Those systems are the ones built from a finite stock of multiplying steps.
Catalog record
| CREATOR | DIMITRIOS CHARAMARAS, ANDREAS MOUNTAKIS AND KONSTANTINOS TSINAS |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 141 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-03504
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On multiplicative recurrence along linear patterns DIMITRIOS CHARAMARAS, ANDREAS MOUNTAKIS AND KONSTANTINOS TSINAS Abstract . In [ 9 ] , Donoso, Le, Moreira and Sun study sets of recurrence for actions of the multiplicative semigroup ( ℕ , × ) (\mathbb{N},\times) and provide some sufficient conditions for sets of the form S = { ( a n + b ) / ( c n + d ) : n ∈ ℕ } S=\{(an+b)/(cn+d){:}\;n\in\mathbb{N}\} to be sets of recurrence for such actions. A necessary condition for S S to be a set of multiplicative recurrence is that for every completely multiplicative function f f taking values on the unit circle, we have that lim inf n → ∞ | f ( a n + b ) − f ( c n + d ) | = 0 . \liminf_{n\to\infty}|f(an+b)-f(cn+d)|=0. In this article, we fully characterize the integer quadruples ( a , b , c , d ) (a,b,c,d) which satisfy the latter property. Our result generalizes a result of Klurman and Mangerel from [ 22 ] concerning the pair ( n , n + 1 ) (n,n+1) , as well as some results from [ 9 ] . In addition, we prove that, under the same conditions on ( a , b , c , d ) (a,b,c,d) , the set S S is a set of recurrence for finitely generated actions of ( ℕ , × ) (\mathbb{N},\times) . 1991 Mathematics Subject Classification Primary: 11N37; Secondary: 37A44. 1. Introduction Since Furstenberg’s multiple recurrence theorem [ 15 ] , which yielded an ergodic proof of Szemerédi’s theorem [ 29 ] , significant progress has been made in understanding recurrence phenomena in dynamical systems generated by a single transformation, that is, for actions of (\mathbb{Z},+) (see for instance [ 4 , 16 , 20 , 28 ] ). Several results have also been established for actions of \mathbb{Z}^{d} [ 1 , 5 , 17 ] , and other amenable groups [ 6 ] . We refer the reader to [ 2 ] for further relevant results and references. Recurrence involving multiplicative actions of the semigroup (\mathbb{N},\times) was first studied in [ 3 ] , but the results in this direction are sparse and several interesting questions remain unanswered. Recently, the connection of this topic to major open conjectures concerning the partition regularity of quadratic equations [ 11 , 12 , 13 ] has sparked renewed interest. In [ 9 ] , Donoso, Le, Moreira and Sun study sets of recurrence, in both measurable and topological settings, for actions of (\mathbb{N},\times) and (\mathbb{Q}_{>0},\times) . In particular, they show [ 9 , Theorem 1.6] that sets of the form S=\{(an+b)/(cn+d){:}\;n\in\mathbb{N}\} , are sets of topological multiplicative recurrence if a=c , and a\mid b or a\mid d . However, there are sets, such as \{(6n+3)/(6n+2){:}\;n\in\mathbb{N}\} , for which they do not provide a conclusive answer, and no counterexample is currently known. This led them to ask the following: Question 1 . [ 9 , Question 7.2] For a\in\mathbb{N} , b,d\in\mathbb{Z} , is it true that \{(an+b)/(an+d){:}\;n\in\mathbb{N}\} is a set of topological multiplicative recurrence if and only if a\mid b or a\mid d ? Recall that a function f:\mathbb{N}\to\mathbb{C} is called multiplicative if f(mn)=f(m)f(n) for all m,n\in\mathbb{N} with (m,n)=1 , and it is called completely multiplicative if the previous identity holds for all m,n\in\mathbb{N} . It can be shown (using a variant of Lemma 3.1 below for topological recurrence) that for a,c\in\mathbb{N} and b,d\in\mathbb{Z} , if the set S defined above is a set of topological multiplicative recurrence, then for any completely multiplicative function f{:}\;\mathbb{N}\to\mathbb{S}^{1} , (1) \liminf_{n\to\infty}|f(an+b)-f(cn+d)|=0. In an attempt to shed some light on Question 1 , we give necessary and sufficient conditions on a,b,c,d so that ( 1 ) holds for every completely multiplicative function f{:}\;\mathbb{N}\to\mathbb{S}^{1} . Note that factoring out the greatest common divisor u=(a,b,c,d) in ( 1 ), we may assume without loss of generality that (a,b,c,d)=1 . Intuitively, if we account for some obvious divisibility obstructions, the multiplicative structure of the numbers an+b and cn+d should be somewhat independent of one another. Thus, for a completely multiplicative function f{:}\;\mathbb{N}\to\mathbb{S}^{1} , one should expect ( 1 ) to hold unless f interacts with the arithmetic (and archimedean) properties of a,b,c,d in a specific way. In fact, we will see that the main counterexamples to ( 1 ) come from a class of multiplicative functions which are essentially of the form \chi(n)n^{it} for some Dirichlet character \chi and a real number t . 1.1. Results on multiplicative functions Let us denote by \mathcal{M} the set (2) \mathcal{M}=\{f{:}\;\mathbb{N}\to\mathbb{S}^{1}{:}\;f\ \text{ is completely multiplicative}\}, where \mathbb{S}^{1} denotes the set of complex numbers with modulus 1 . The first theorem we prove in this paper is the following: Theorem 1.1 . Let a,c\in\mathbb{N} , b,d\in\mathbb{Z} with (a,b,c,d)=1 . Then, the following are equivalent: (1) For every f\in\mathcal{M} and \varepsilon>0 , the set (3) \mathcal{A}(f,\varepsilon):=\{n\in\mathbb{N}{:}\;|f(an+b)-f(cn+d)|<\varepsilon\} has positive upper logarithmic density. 1 1 1 See Section 2 for definition. (2) For every f\in\mathcal{M} , \liminf_{n\to\infty}|f(an+b)-f(cn+d)|=0. (3) a=c , and b=d or a\mid bd . Remark 1 . We note here that if f is pretentious or finitely generated (see Definitions 2.1 and 2.4 ), then the set \mathcal{A}(f,\varepsilon) above has positive lower logarithmic density. We prove this in Section 5 . For instance, Theorem 1.1 implies that for all f\in\mathcal{M} , \liminf_{n\to\infty}|f(6n+3)-f(6n+2)|=0. This provides positive evidence that \{(6n+3)/(6n+2){:}\;n\in\mathbb{N}\} is a set of multiplicative recurrence. In [ 22 , Theorem 1.1] , Klurman and Mangerel prove that for any f\in\mathcal{M} , \liminf_{n\to\infty}|f(n+1)-f(n)|=0. Theorem 1.1 generalizes this result, as in place of (n+1,n) we can put any pair (an+b,an+d) with au\mid bd , where u=(a,b,d) , and our conditions on a,b,d are optimal. Furthermore, Theorem 1.1 generalizes [ 9 , Corollary 1.7] . We also prove the following result concerning pairs of multiplicative functions. Theorem 1.2 . For any a\in\mathbb{N} , f,g\in\mathcal{M} and \varepsilon>0 , the set \mathcal{A}(f,g,\varepsilon):=\{n\in\mathbb{N}{:}\;|f(an+1)-g(an)|<\varepsilon\} has positive upper logarithmic density. In particular, this implies that for any a\in\mathbb{N} and f,g\in\mathcal{M} \liminf_{n\to\infty}|f(an+1)-g(an)|=0. Theorem 1.2 also generalizes [ 22 , Theorem 1.1] , as it holds for pairs f,g in \mathcal{M} , and (n+1,n) can be replaced by (an+1,an) for any a\in\mathbb{N} . Remark 2 . In Section 6 we show the existence of a pair f,g\in\mathcal{M} that satisfy the property \liminf_{n\to\infty}|f(n+2)-g(n)|>0 . Therefore, even for a=1 in the context of Theorem 1.2 , one cannot replace (n+1,n) by (n+k,n) for any k\in\mathbb{Z} . Hence, it seems difficult to prove an analog of Theorem 1.1 for pairs f,g\in\mathcal{M} , unless one restricts to a smaller class of multiplicative functions. 1.2. Multiplicative recurrence for finitely generated systems Motivated by [ 9 , Theorem 1.6 and Question 7.2] , where the authors investigate under what conditions on a,b,c,d the set S=\{(an+b)/(cn+d){:}\;n\in\mathbb{N}\} is a set of topological multiplicative recurrence, we are interested in finding sufficient conditions so that S is a set of measurable multiplicative recurrence (see Definition 1.4 ) (a property stronger than the topological one). In view of Theorem 1.1 , it is reasonable to ask whether the condition a=c , and b=d or a\mid bd is sufficient to show that S is a