A Szemeredi-type regularity lemma in abelian groups, with applications

TL;DR

This paper establishes a Szemeredi-type regularity lemma for abelian groups and applies it to additive number theory, providing structural results for almost sum-free sets and progressions in dense subsets.

Contribution

It introduces a novel regularity lemma for abelian groups and uses it to derive new results in additive combinatorics, including structure theorems and progressions in dense sets.

Findings

Structural decomposition of almost sum-free sets

Existence of non-zero common difference in dense subsets with many 3-term progressions

Extension of regularity methods to abelian groups

Abstract

Szemeredi's regularity lemma is an important tool in graph theory which has applications throughout combinatorics. In this paper we prove an analogue of Szemeredi's regularity lemma in the context of abelian groups and use it to derive some results in additive number theory. One is a structure theorm for sets which are almost sum-free. If A is a subset of [N] which contains just o(N^2) triples (x,y,z) such that x + y = z then A may be written as the union of B and C, where B is sum-free and |C| = o(N). Another answers a question of Bergelson, Host and Kra. If alpha, epsilon > 0, if N > N_0(alpha,epsilon) and if A is a subset of {1,...,N} of size alpha N, then there is some non-zero d such that A contains at least (alpha^3 - epsilon)N three-term arithmetic progressions with common difference d.