No label defined (Q4256478)

Let \(A\) be a set of \(k\) residues modulo a prime \(p\). About 40 years ago Erdős and Heilbronn conjectured that the number of residues representable in the form \(a+b\) with \(a,b\in A\), \(a\neq b\), is at least \( \min(2k-3, p)\). This conjecture is established under the assumption that \(60<k<p/50\). Furthermore, the authors describe the structure of those sets, for which the number of such sums is \(<2.006k\). They show that in this case the set is contained in a relatively short arithmetic progression. Between the submission and publication of this paper (these are the proceedings of a 1993 conference) the Erdős-Heilbronn conjecture has been fully established by \textit{J. A. Dias da Silva} and \textit{Y. O. Hamidoune} [Bull. Lond. Math. Soc. 26, 140-146 (1994; Zbl 0819.11007)]. A different proof and an extension to several distinct summands was given by \textit{N. Alon, M. B. Nathanson} and \textit{I. Z. Ruzsa} [Am. Math. Mon. 102, 250-255 (1995; Zbl 0849.11081); J. Number Theory 56, 404-417 (1996; Zbl 0861.11006)]. These proofs, however, cannot yield such an extension to the case when the number of restricted sums is just above the minimal possible value. For the entire collection see [Zbl 0919.00044].