``Poly-unsaturated'' posets: The Greene-Kleitman theorem is best possible (Q1069961)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | ``Poly-unsaturated'' posets: The Greene-Kleitman theorem is best possible |
scientific article |
Statements
``Poly-unsaturated'' posets: The Greene-Kleitman theorem is best possible (English)
0 references
1986
0 references
Let \(P_ 3\) be the poset \(\{a>b>c>d\), \(b>e\), \(f>c\}\). Suppose that \(P_ 3,\ldots,P_{n-1}\) has been constructed. In order to construct \(P_ n\) take two copies of \(P_{n-1}\), a left or lower copy \(P^-_{n-1}\) and a right or upper copy \(P^+_ n\). Insert a chain \(\{a_ n>b_ n>u_ 3>\cdots >u_{n-1}>c_ n>d_ n\}\) and label the corresponding elements in \(P^-_{n-1}\) as \(\{a^-_ n>b^-_ n>u^-_ 3>\cdots >u^- _{n-2}>c^-_ n>d^-_ n\}\) and in \(P^+_ n\) as \(\{a^+_ n>b^+_ n>u^+_ 3>\cdots >u^+_{n-2}>c^+_ n>d^+_ n\}.\) Set \(b_ n>b^-_ n\) and \(c^+_ n>c_ n\) and use transitivity to extend the construction. The posets \(P_ n\) are rather remarkable posets as is demonstrated in this paper. Greene and Kleitman proved that for any poset P and any integer k, P has a chain partition that is both k- and \(k+1\)-saturated. The poset \(P_ n\) has height precisely n and the property that no chain partition of \(P_ n\) is k-saturated for any two non-consecutive values of \(k\leq n\). This demonstrates that the Greene-Kleitman Theorem cannot be improved upon in general. The author of this very worthwile paper also provides a construction of \(P_ n\) as the intersection of two linear extensions \(L_ 1(P_ n)\) and \(L_ 2(P_ n)\), thereby demonstrating that \(P_ n\) is 2- dimensional. If \(Q_ n\) is the poset formed by inverting \(L_ 2(P_ n)\) and intersecting it with \(L_ 1(P_ n)\), then a poset is obtained in which elements are related if and only if they are unrelated in \(P_ n\). Thus, \(Q_ n\) is a poset of width \(n+1\), with no antichain partition that is k-saturated for nonconsecutive values of \(k\leq n\). Q has a completely saturated chain partition just as \(P_ n\) has a completely saturated antichain partition. The results obtained in this paper support once more the dual of the cosmologists' principle which seems to obtain in poset theory: if a particular but general conclusion is not absolutely permitted, then there is some poset or class of posets which forbids that it be made. For the hunter of counterexamples this is an optimists' principle to go by.
0 references
posets
0 references
height
0 references
Greene-Kleitman Theorem
0 references
linear extensions
0 references
width
0 references
completely saturated chain partition
0 references
completely saturated antichain partition
0 references
counterexamples
0 references