An ordinal partition avoiding pentagrams (Q4527907)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1558169
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | An ordinal partition avoiding pentagrams |
scientific article; zbMATH DE number 1558169 |
Statements
An ordinal partition avoiding pentagrams (English)
0 references
28 August 2001
0 references
ordinary partition relations
0 references
countable ordinals
0 references
ordinal exponentiation
0 references
additively indecomposable ordinals
0 references
0.8377026915550232
0 references
0.8194848895072937
0 references
0.796867311000824
0 references
For ordinal numbers \(\alpha, \beta\) and \(\gamma\), the ordinary partition relation symbol \(\alpha \to (\beta, \gamma)^2\) means that whenever the unordered pairs from \(\alpha\) are colored with two colors, say red and blue, then either there is a subset \(X\) of \(\alpha\) of order type \(\beta\) such that all pairs from \(X\) are red, or there is a subset \(Y\) of \(\alpha\) of order type \(\gamma\) such that all pairs from \(Y\) are blue. The negation of this statement is written \(\alpha \not \to (\beta, \gamma)^2\). In this language, Ramsey's Theorem says \(\omega \to (\omega, \omega)^2\), and Specker, Chang, and Milner proved that \(\alpha \to (\alpha, n)^2\) for all finite \(n\) in case \(\alpha = \omega^2\) or \(\omega^\omega\). On the other hand, Galvin showed that if \(\beta \geq 3\) is not AI, then \(\omega^\beta \not\to (\omega^\beta , 3)^2\). (Here \(\beta\) is AI (additively indecomposable) if it is not the sum of two smaller ordinals iff either \(\beta = 0\) or \(\beta = \omega^\gamma\) for some \(\gamma\).) These results were all known by 1975. Using an intricate analysis of certain colorings of pairs of finite increasing sequences from \(\omega\), the author proves that if \(\beta\) is not AI, then \(\omega^{\omega^\beta} \not\to (\omega^{\omega^\beta}, 5)^2\). The proof uses methods due to \textit{C. Darby} [J. Comb. Theory, Ser. B 76, No. 2, 205-222 (1999; Zbl 0935.03054)] and improves on the latter's earlier \(\omega^{\omega^\beta} \not\to (\omega^{\omega^\beta}, 6)^2\) for \(\beta\) not AI (a result obtained independently by \textit{R. Schipperus} [Countable partition ordinals (Ph.D. thesis, Boulder) (1999)]). On the positive side, recent known results are \(\omega^{\omega^\beta} \to (\omega^{\omega^\beta}, 3)^2\) for \(\beta\) AI or \(\beta = \gamma + \delta\) with \(\gamma, \delta\) AI (Schipperus), and \(\omega^{\omega^2} \to (\omega^{\omega^2}, 4)^2\) (Darby and Larson). This means that for \(\beta = 2\), the result of the paper is optimal.
0 references