Skip to content

Erdős Problem #30

Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: open (not automatically the live result).
Subject areas: number theory, sidon sets, additive combinatorics.

Read this status correctly

The status above is reproduced from GCL's immutable external-catalogue snapshot. It is not a new claim of current mathematical resolution, a source-semantic review, an independent proof check, or GCL certification. A published proof and a GCL replay are separate events.

The question in plain language

A Sidon set is a set of integers whose pairwise sums are unique up to swapping the summands, including sums with a repeated summand. Let \(h(N)\) be the largest size of such a set contained in \(\{1,\ldots,N\}\). Erdős Problem #30 asks whether, for every \(\varepsilon>0\),

\[ h(N)=\sqrt{N}+O_{\varepsilon}(N^\varepsilon). \]

The quantifier matters: the error must be smaller than every fixed positive power of \(N\), not merely bounded by a single power such as \(N^{1/4}\).

A small example

The set \(A=\{1,2,4\}\) is Sidon: its six unordered sums with repetition are \(2,3,5,4,6,8\), all distinct. Thus \(h(4)\ge3\). For a Sidon set of size \(k\), the \(k(k+1)/2\) unordered pair sums are different integers from 2 through \(2N\); therefore

\[ \frac{k(k+1)}2\le2N-1. \]

This elementary finite bound is much weaker than the conjectured asymptotic precision.

What is known

The maintained registry reports a sharper upper bound of the form \(\sqrt N+0.98183N^{1/4}+O(1)\) from Carter, Hunter and O'Bryant (2025), while the exact desired \(O_\varepsilon(N^\varepsilon)\) error remains open. The source audit did not independently inspect the full modern paper, so the numerical coefficient is attributed to the registry rather than newly source-certified here.

GCL's current state

GCL audit #1299 reconciles the repeated-summand interpretation with the pinned formal Sidon predicate, establishes the elementary pair-counting inequality and reports exhaustive integer enumeration of \(h(N)\) for \(1\le N\le12\). The pinned proof body is unfilled, and none of these finite calculations establishes the asymptotic conjecture.

What remains

One needs a construction and upper estimate pinning the maximal Sidon size down to square-root scale with subpolynomial error for every positive exponent tolerance. Pair counting by itself stops at a substantially weaker leading constant and is not enough.

Sources and claim boundaries

Catalogue and evidence provenance

  • Pinned catalogue source: teorth/erdosproblems at 536408716724.
  • GCL issue: #1299 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
  • GCL triage: PROTECTED_READY_BACKLOG.
  • Formal-statement coverage: A provider statement was included in the pinned snapshot. Presence does not mean the statement or proof has been checked.
  • External mathematical status: consult current problem page and its dated bibliography.

Back to catalogue · Compendium introduction