Erdős Problem #152
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: proved (not automatically the live result).
Subject areas: sidon sets.
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 finite set \(A\) of natural numbers is a Sidon set if each sum \(a+b\) with \(a,b\in A\) and \(a\le b\) identifies the unordered pair uniquely. Form the sumset \(S=A+A\) and call \(s\in S\) isolated when neither \(s-1\) nor \(s+1\) belongs to \(S\).
Erdős, Sárközy and Sós asked whether, for every fixed positive integer \(M\), every sufficiently large finite Sidon set has at least \(M\) isolated elements in \(A+A\). The stronger conjecture asks whether the number of isolated sums is bounded below by a constant times \(|A|^2\).
A small example
The set \(A=\{1,2,4\}\) is Sidon: its unordered sums (allowing equal terms) are exactly \(2,3,4,5,6,8\), all distinct. Among these, \(8\) is isolated because neither \(7\) nor \(9\) is a sum; the other five sums have a neighbouring element of the sumset. Thus this example has one isolated sum.
An isolated element is counted once, not once for each representation as \(a+b\). For a Sidon set, sums with \(a\ne b\) have a unique unordered representation, but \(a+a\) is still part of \(A+A\). Omitting diagonal sums would produce the wrong isolation test.
What is known
PROVED, including a quadratic lower bound. A Google DeepMind prover agent produced an external formal proof of the original assertion and a stronger numerical inequality. As recorded in the maintained Erdős #152 discussion and upstream Formal Conjectures entry, if \(I(A+A)\) is the number of isolated sums and \(n=|A|\), then
so for large \(n\), \(I(A+A)\gg n^2\). This explicitly implies the original question for arbitrarily large fixed \(M\) by taking \(n\) sufficiently large.
The source records two exact third-party formal proof revisions: the original positive-growth proof and the stronger quadratic variant. GCL has not independently executed those exact Lean toolchains or audited their complete axiom dependencies. A related extension to truncations of infinite Sidon sets is a separate assertion requiring its own hypotheses.
GCL's current state
The pinned and maintained catalogue classify #152 as PROVED. The original issue is #1421. The public primary proof pointer and exact constant-bearing inequality were checked, but proof status is externally reported, not protected GCL formal admission.
The original issue contains an authenticated RESULT/1 source report. This is unadjudicated source evidence only, not independent mathematical certification or editorial admission.
What remains
A bounded task enumerates all Sidon subsets of \(\{1,\ldots,12\}\), forms \(A+A\) including diagonal terms, and counts isolated values exactly. Acceptance: the example \(\{1,2,4\}\) yields one isolated sum, every unordered sum is unique, and the inequality above is tested as a finite consistency condition without treating numerical tests as its proof. Formal source review should replay the precise second external Lean snapshot with axiom and theorem matching.
Sources and claim boundaries
- Canonical Erdős #152, current catalogue statement and discussion.
- DeepMind source discussion, Erdős #152.
- Formal proof of the quadratic variant, revision ff58c933.
- GCL original source issue #1421.
No new Sidon theorem, independent Lean kernel proof, or protected editorial admission is claimed.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #1421 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: OUTSIDE_STRICT_OPEN_TRIAGE.
- 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.