The last exception for two-colour Rado numbers
- Kind
- Theorem, a written proof with no computer search
- Status
- Proved and Lean-verified for every (both upper and lower halves in core Lean 4); unrefereed
- Prior work
- Not found in the searches logged; answers Kim, Song and Hwang (2023); the lower half is Martinelli and Schaal's Lemma 3
On this page
The statement
Colour each number from 1 to red or blue. A solution is three numbers from 1 to , repeats allowed, with . The Rado number is the least for which every colouring of has a solution all of one colour. It exists exactly when is valid ( odd or even). Martinelli and Schaal (2007) proved the lower bound
An exception is a valid with . Kim, Song and Hwang (Discrete Mathematics, 2023) proved there are none once (for ), defined as the least from which for every valid , and wrote:
"It would also be interesting to determine ν(k) for all k. ... We conjecture that ν(5) = 66."
Theorem. For every the last exception is
, and for every valid . So for odd and for even (in the paper's reading of ).
| 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | |
|---|---|---|---|---|---|---|---|---|---|---|
| 12 | 28 | 65 | 102 | 182 | 248 | 387 | 490 | 704 | 852 | |
| 13 | 30 | 66 | 104 | 183 | 250 | 388 | 492 | 705 | 854 | |
| printed bound | 295 | 498 | 777 | 1144 | 1611 | 2190 | 2893 | 3732 |
, the printed conjecture, is the case . The printed bound is too large by a factor that tends to 4.
Check it yourself
- , : . Colouring 1–4 blue and 5–7 red is good on . On , say 1 is blue; then every colour is forced: makes 5 red, makes 7 red, makes 3 blue, makes 4 blue, makes 6 red, makes 2 blue, makes 8 blue, and then is all blue.
- , : the last exception. The colouring blue, red, red, blue, red, blue, blue, red of has no one-colour solution among its 11 solutions (figure 1), so , one more than .
- : . Kim, Song and Hwang's own Example is a good colouring of for , so ; from on, .
Worked examples
How the proof goes. Replacing every in by turns the constant into ; at and the equation becomes with . This reflection is the system's own: a reviewer agent gave it in an earlier round of the same system's work. The proof then follows forced colours, as in the hand case, through four lemmas to a one-colour solution on the needed stretch. An arithmetic lemma shows the stretch is long enough for every valid , and fails by exactly one at .
Status and who checked it
- Lean-verified for every (both halves). Both the upper
half (no exception above ) and the lower half ( is an
exception) are proved unconditionally in the core Lean 4 archive
(
lean/RadoNu/RadoNu.lean, zerosorry, standard axioms only). Before formalization, the written proof of the upper half was re-derived line by line by a falsifier agent ("No gap found") and a reviewer agent; no human referee has read it. It does not use the 2023 paper's threshold theorem. - Computer checks, not the proof: programs replay every step of the proof's chains for up to 64 with 0 gaps; SAT searches find no exception just above for to 18 and 31 to 60; a second exhaustive solver settles for to 16 by a third route, up to the printed threshold.
- Open: most values below , which follow no single formula.
- Unrefereed.
What is known, and not credited here. The lower bound and, in their Lemma 3, the lower half for every are Martinelli and Schaal's (2007); is Kézdy, Snevily and White's (2009) and follows from Guo (2010) and the 2023 Remark. Jones and Schaal's Lemma 1 (2004) is the translation step, which shifts every number of a colouring by ; the written proof uses it for a companion result on smaller , not for the reflection. The claim here is the upper half for every .
Prior work
Searches run. A falsifier agent and a reviewer agent logged 33 and 26 searches and read 22 and 20 candidates, with hand searches of DataCite, Zenodo, HAL, OSF, GitHub, citing works and the web. Two expository surveys of 2026 state only the 2023 threshold theorem. A keyed CORE run on 2026-11-20 (7 requests answered, 71 records) found two records on the same equation with other claims: Kim, Hwang and Song (2022), on small odd , and an Indonesian thesis of 2025 (Fikri Rahman D.) on and , below every last exception there. No record determines or gives the last exception for any . The result was not found in the searches logged.
Still owed: Google Scholar and its cited-by list of Kim, Song and Hwang (2023); the full texts of Jones and Schaal (2004), Kim, Hwang and Song (2022), Bergjord's 2023 thesis and the 2025 Indonesian thesis; Landman and Robertson's Ramsey Theory on the Integers; MathSciNet.
Further reading
- Kim, Song and Hwang, "Two-color Rado number of x + y + c = kz for large c", Discrete Mathematics 346 (2023) 113597
- Jones and Schaal, "Two-color Rado numbers for x+y+c=kz", Discrete Math. 289 (2004)
- Martinelli and Schaal, Ars Combin. (2007), the lower bound.
- Kézdy, Snevily and White (2009), the case ; Guo (2010), the case .