Skip to content

Erdős problem 835

Does there exist a k>2k>2 such that the kk-sized subsets of {1,...,2k} can be coloured with k+1k+1 colours such that for every A⊂{1,…,2k}A\subset \{1,\ldots,2k\} with ∣A∣=k+1\lvert A\rvert=k+1 all k+1k+1 colours appear among the kk-sized subsets of AA?

Sources

The source's own material, retained exactly.

13 retained statements36c6d0fef71a

Open selected source

FormalConjectures/ErdosProblems/

835.lean

Retained formal statement8 of 13

Ma and Tang have proved that the chromatic number of J(2k,k)J(2k,k) is >k+1>k+1 for all k>2k>2 not of the form p−1p-1 for prime pp.

FormalConjectures/ErdosProblems/835.leanErdos835.johnson_chromaticNumber_composite1 lineExact file
∀ (k : ℕ), 2 < k → (k + 1).Composite → ↑k + 1 < (SimpleGraph.johnson (2 * k) k).chromaticNumber
SolvedStatement only, no proof