English

A Roth type theorem for dense subsets of $\mathbb{R}^d$

Combinatorics 2017-06-07 v2 Classical Analysis and ODEs Number Theory

Abstract

Let 1<p<1 < p < \infty, p2p\neq 2. We prove that if ddpd\geq d_p is sufficiently large, and A\subsRdA\subs\R^d is a measurable set of positive upper density then there exists \la0=\la0(A)\la_0=\la_0(A) such for all \la\la0\la\geq\la_0 there are x,yRdx,y\in\R^d such that {x,x+y,x+2y}\subsA\{x,x+y,x+2y\}\subs A and yp=\la|y|_p=\la, where yp=(iyip)1/p||y||_p=(\sum_i |y_i|^p)^{1/p} is the lp(Rd)l^p(\mathbb R^d)-norm of a point y=(y1,,yd)Rdy=(y_1,\ldots,y_d)\in\R^d. This means that dense subsets of Rd\R^d contain 3-term progressions of all sufficiently large gaps when the gap size is measured in the lpl^p-metric. This statement is known to be false in the Euclidean l2l^2-metric as well as in the l1l^1 and \ell^{\infty}-metrics. One of the goals of this note is to understand this phenomenon. A distinctive feature of the proof is the use of multilinear singular integral operators, widely studied in classical time-frequency analysis, in the estimation of forms counting configurations.

Keywords

Cite

@article{arxiv.1511.06010,
  title  = {A Roth type theorem for dense subsets of $\mathbb{R}^d$},
  author = {Brian Cook and Ákos Magyar and Malabika Pramanik},
  journal= {arXiv preprint arXiv:1511.06010},
  year   = {2017}
}
R2 v1 2026-06-22T11:48:57.813Z