Classification theorems for sumsets modulo a prime (Q1024349)

Let \(A\) be a (multi-)subsequence of \(\mathbb{F}_{p}\) with \(p\) a prime. This paper addresses the following questions: (1) When can 0 be represented as a sum of elements of \(A\)? (2) When can every element of \(\mathbb{F}_{p}\) be represented as a sum of elements of \(A\)? It is straightforward to see that if the elements of \(A\) add up to a number less than \(p\), then 0 cannot be represented as a sum of elements of \(A\) (we say \(A\) is \textit{zero-sum-free}). Theorem 2.2 shows that the converse of this statement holds in approximate form. That is, if \(A\) is zero-sum-free then \(A\) must essentially (after dilating and possibly removing a small number of elements of \(A\)) be such that the sum of its elements is less than \(p\). Similarly, it is not difficult to establish that if \(\sum_{a \in A}\|a\| <p\), where \(\|x\|\) denotes the distance of \(x\) from 0, then not every element of \(\mathbb{F}_{p}\) can be represented as the sum of elements of \(A\) (we say \(A\) is \textit{incomplete}). Theorem 2.5 gives the reverse direction of this characterization: if \(A\) is incomplete, then it must contain a dense subsequence the norms of whose elements (after a possible dilation) add up to less than \(p\). An analogous statement is formulated for the second problem in the case where the sums under consideration consist of a fixed number of elements (Theorem 2.8). The main tool in the proof of these results is a theorem from [\textit{E. Szemerédi} and \textit{V. Vu}, ``Long arithmetic progressions in sumsets: thresholds and bounds'', J. Am. Math. Soc. 19, No. 1, 119--169 (2006; Zbl 1088.11012)], which states that provided that \(A\) is sufficiently large, the set of sums of precisely \(l\) elements of \(A\) contains a long arithmetic progression. Amongst the many applications of the classification theorems in this paper are a refinement of the Erdős-Ginzburg-Ziv theorem (Theorem 6.3) first obtained by Peterson and Yuster (see Section 7 of [\textit{A. Geroldinger}, Additive group theory and non-unique factorizations, in: Combinatorial number theory and additive group theory, in: Advanced Courses in Mathematics -- CRM Barcelona, Basel: Birkhäuser, 1--86 (2009; Zbl 1221.20045)]), and results regarding the number of zero-sum-free and incomplete subsets of \(\mathbb{F}_{p}\) (Corollaries 5.3 and 5.5). The paper is clearly structured and written and therefore very accessible.