Wiki
Wiki

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

Updated


Claim. For every ϵ>0\epsilon>0 and every integer t≥1t\ge1 there is N0N_0 such that, whenever N≥N0N\ge N_0, every A⊆[t]NA\subseteq[t]^N with ∣A∣≥ϵtN|A|\ge\epsilon t^N contains a combinatorial line: the density Hales--Jewett theorem, which is the question of Problem 171, answered yes. P. Dodos, V. Kanellopoulos and K. Tyros, A simple proof of the density Hales--Jewett theorem, Int. Math. Res. Not. IMRN 2014, no. 12, 3340--3352, doi:10.1093/imrn/rnt041, arXiv:1209.4986 (v1 22 September 2012, v2 19 March 2013). The paper gives a purely combinatorial proof modeled on the Polymath density-increment argument (claim page) but shorter: it avoids the equal-slices measure and works with the uniform measure throughout. The paper is not held in the library and is not among the site's references; its statement is taken from its abstract and from the Lean file below, and the theorem itself is the one of Furstenberg and Katznelson (claim page).

Depends on. No page of this wiki: the proof is self-contained, a third route to the theorem beside the Furstenberg--Katznelson and Polymath proofs.

Acceptance. Refereed: the paper appeared in International Mathematics Research Notices, published online 15 March 2013 and in print in the 2014 volume, issue 12, as the publication record dates it. Not reviewed: the site's curator credits the problem to Furstenberg and Katznelson and to the Polymath project, not to this paper, and no outside reviewer of this proof is documented. Nothing here is this project's own review of the proof.

Formalization. The file src/latest/ErdosProblems/Erdos171.lean of Boris Alexeev's lean-proofs repository (first added 2026-08-17, last changed 2026-08-24, pinned at the commit of 2026-09-15) declares itself a formalization of a solution to the problem: its header lists Dodos, Kanellopoulos and Tyros as informal authors and Codex and GPT-5.6 Sol as formal authors, and its module comment says the proof follows their uniform-measure density-increment argument, with the Hales--Jewett theorem, a derived line-coloring Graham--Rothschild theorem, Sperner's theorem for the binary base case, uniform-fibre regularization, structured correlation by insensitive sets and a greedy subspace tiling as its principal inputs. It proves Erdos171.erdos_171: for every real ϵ>0\epsilon>0 and every t≥1t\ge1 there is N0N_0 such that every Finset (Word t N) of cardinality at least ϵtN\epsilon t^N with N≥N0N\ge N_0 satisfies the development's own ContainsLine; it closes with #print axioms without the printed output. The formal-conjectures statement file for the problem, which states the question with Mathlib's Combinatorics.Line, is tagged solved at its commit of 2026-10-06 and names this file as the formal proof. The community database (teorth/erdosproblems) lists, the problem as "proved (Lean)" with formal_status Lean, both as of their last update on 2026-08-24, and formalized "yes" as of its last update on 2026-09-20, without dating when either state changed. This corpus has not built or audited the development, and the fidelity of its definitions to the site's question has not been independently reviewed, so the page lists no formalized evidence.