Wiki
Wiki

Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.

Updated

Problem 1025

../

claims/: The 4 claim pages of Problem 1025, one per claimant's result; the problem's standing derives from them.


Statement. Let ff be a function from all pairs of elements in {1,…,n}\{1,\ldots,n\} to {1,…,n}\{1,\ldots,n\} such that f(x,y)≠xf(x,y)\neq x and ≠y\neq y for all x,yx,y. We call X⊆{1,…,n}X\subseteq \{1,\ldots,n\} independent if whenever $x,y\in X$ we have f(x,y)∉Xf(x,y)\not\in X.

Let g(n)g(n) be such that, in every function ff, there is an independent set of size at least g(n)g(n). Estimate g(n)g(n).

Status. The site labels the problem SOLVED (LEAN), the Lean qualification recorded in the community database since 2026-09-15, crediting Spencer [Sp72] with g(n)≫n1/2g(n)\gg n^{1/2} and Conlon, Fox and Sudakov [CFS16] with g(n)≪n1/2g(n)\ll n^{1/2}, so g(n)≍n1/2g(n)\asymp n^{1/2}.

Source. erdosproblems.com/1025, accessed 2026-09-04. Cite as: T. F. Bloom, Erdős Problem #1025, https://www.erdosproblems.com/1025.

References.

  • [CFS16] Conlon, David and Fox, Jacob and Sudakov, Benny, Short proofs of some extremal results II. J. Combin. Theory Ser. B (2016), 173-196.
  • [ErHa58] Erdős, P. and Hajnal, A., On the structure of set mappings. Acta Math. Acad. Sci. Hungar. 9 (1958), 111-131.
  • [Fu91] Füredi, Z., Maximal independent subsets in Steiner systems and in planar sets. SIAM J. Discrete Math. 4 (1991), 196-199.
  • [Sp72] Spencer, Joel, Turán's theorem for kk-graphs. Discrete Math. (1972), 183-186.

Formalization. Statement in formal-conjectures, marked solved there as of its commit of 19 September 2026 and pointing at a third-party Lean proof, linked from the claim page of Conlon, Fox and Sudakov, which this corpus has not built.

Current assessment

The question asks for the order of g(n)g(n), the largest independent set that every mapping of pairs of {1,…,n}\{1,\ldots,n\} to points outside the pair must admit; it is p(n,2,1)p(n,2,1) in the notation of Conlon, Fox and Sudakov (p(n,1,2)p(n,1,2) in Erdős and Hajnal's Theorem 12). Erdős and Hajnal's paper On the structure of set-mappings posed the question and gave n1/3≪g(n)≪(nlog⁡n)1/2n^{1/3}\ll g(n)\ll(n\log n)^{1/2}, the accepted partial claim on Erdős and Hajnal 1958, both bounds since superseded. The order is n1/2n^{1/2}: the lower bound is Spencer's Turán theorem for hypergraphs, applied to the triples {x,y,f(x,y)}\{x,y,f(x,y)\}, on the accepted partial claim page Spencer 1972, and the upper bound is the grid construction of Conlon, Fox and Sudakov, Section 2 of Short proofs of some extremal results II, on the accepted full claim page Conlon, Fox and Sudakov 2016, which states the two-sided Θ(n1/2)\Theta(n^{1/2}) and rests on Spencer's page. Both are refereed journal articles credited by the site's curator. The paper of Conlon, Fox and Sudakov records that the upper bound g(n)≪n1/2g(n)\ll n^{1/2} was proved independently and much earlier by Füredi [Fu91], Theorem 2.3 of Maximal independent subsets in Steiner systems and in planar sets, which proves (23/9)n<g(n)<2n(2\sqrt3/9)\sqrt n<g(n)<2\sqrt n with the lower bound from Spencer's theorem; the site does not cite Füredi, and his two-sided theorem is a second accepted full claim, on the page Füredi 1991. The accepted claim credited by the site is Conlon, Fox and Sudakov's, which rests on the accepted partial claim of Spencer. The asymptotic constant is open. Two third-party Lean proofs of the two-sided estimate, one in Boris Alexeev's repository and one released by IIIS Lean, which the community database cites as the problem's formalization, are linked from the claim pages of Conlon, Fox and Sudakov and of Spencer; the corpus has built neither. The site's page listed no comment or proof claim.

Linked library material

These entries are derived from explicit links on library pages. They are navigation only and do not by themselves record mathematical progress.