3
$\begingroup$

Suppose $A\subseteq \mathbb{Z}$ has positive upper density, that is $$ \limsup_{n\to \infty} \frac{|A\cap\{1,2,\dots,n\}|}{n}>0. $$ I would like to know whether the following statement for $A$ can be proven without applying Szemerédi's theorem on arithmetic progression:

For any integer $n$, there exists an integer $b$ such that $\{b, 2b,\dots, nb\}\subseteq A-A$, where $A-A=\{a_1-a_2: a_1, a_2\in A\}$.

$\endgroup$
3
  • $\begingroup$ Isn't $A-A$ empty? $\endgroup$
    – Mike Stay
    Jan 26, 2017 at 19:51
  • 1
    $\begingroup$ $A - A = \{x \in A : x = y - z \text{ for some } y, z \in A\}$. $\endgroup$ Jan 26, 2017 at 20:02
  • 3
    $\begingroup$ I would expect that it should be possible to modify appropriately the argument of Croot, Ruzsa, and Schoen (people.math.gatech.edu/~ecroot/kterm.pdf) who proved that if $A\subset[1,n]$ has size $|A|>(3n)^{1-1/(k-1)}$, then the sumset $A+A$ contains an arithmetic progression of length at least $k$. $\endgroup$
    – Seva
    Jan 26, 2017 at 20:13

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.