Erdős Problem #20
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: open (not automatically the live result).
Subject areas: 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 sunflower is a collection of sets whose pairwise intersections are all the same common set, called its core. For example, three different sets might all contain one central element and otherwise be disjoint.
Erdős Problem #20 is the famous sunflower conjecture. Fix a number of petals \(k\). Is there a constant \(c_k\) depending on \(k\) but not on the set size \(n\) such that every sufficiently large family of sets each of size \(n\) must contain a sunflower with \(k\) petals once it has roughly \(c_k^n\) members? The fixed-petal-count conjecture remains open.
A small example
The three sets
form a sunflower whose common core is \(\{1\}\). By contrast, three sets \(\{1,2\}\), \(\{2,3\}\) and \(\{1,3\}\) have three different pairwise intersections and hence do not form a sunflower.
For two petals every pair of different sets trivially forms a sunflower. The difficult part begins at fixed \(k\ge3\) and growing set size \(n\).
What is known
The classical Erdős–Rado sunflower lemma gave an upper threshold comparable to \(n^n\) (with constants depending on \(k\)). Alweiss, Lovett, Wu and Zhang, Improved bounds for the sunflower lemma, Annals of Mathematics 194 (2021), 795–815, improved this dramatically to a threshold of order \((\log n)^n\) for fixed \(k\), up to appropriate fixed-parameter factors.
But \((\log n)^n\) cannot be bounded by \(c_k^n\) using one fixed \(c_k\) for all large \(n\). Thus the celebrated improvement is not the conjectured exponential bound.
GCL's current state
Source audit #1289 reads the frozen formal definition of the sunflower threshold, identifies the quantifier requiring \(c_k\) independent of \(n\), and checks trivial one- and two-petal examples. Both the main conjecture and the source's factorial-bound variant have unfilled proof bodies; a verified trivial test case is not a global solution.
What remains
Prove a fixed-base exponential bound for each fixed petal count, or produce a family showing that no such bound can hold. The main obstacle is eliminating the growing \(\log n\) factor from the base of the best published general estimates.
Sources and claim boundaries
- Erdős Problems #20: open fixed-petal formulation.
- Alweiss–Lovett–Wu–Zhang (2021), publisher article, DOI 10.4007/annals.2021.194.3.5: improved bound, not proof of the conjecture.
- GCL source audit #1289: pinned quantifiers and finite examples, not mathematical certification.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #1289 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: DEFER_UNTIL_NEW_MECHANISM.
- 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.