A new proof of a theorem of Harper on the Sperner-Erdős problem (Q1065006)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: A new proof of a theorem of Harper on the Sperner-Erdős problem |
scientific article; zbMATH DE number 3920475
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | A new proof of a theorem of Harper on the Sperner-Erdős problem |
scientific article; zbMATH DE number 3920475 |
Statements
A new proof of a theorem of Harper on the Sperner-Erdős problem (English)
0 references
1985
0 references
Let P be a finite poset. A weight on P is a function \(v: P\to \{x\in R\); \(x>0\}\), a rank on P is a function \(r: P\to \{0,1,2,...\}\) such that \(r(x)=0\), if x is minimal, and \(r(x)=r(y)+1\), if \(x>y\) (i.e., x covers y). For any subset \(S\subseteq P\) define \(v(S)=\sum_{x\in S}v(x)\). Put \(N_ i=\{x\in P;\quad r(x)=i\},\) \(r(P)=\max \{r(x);\quad x\in P\}.\) The weighted and ranked poset P is called normal, if \(v(A)/v(N_ i)\leq v(R(A))/v(N_{i+1})\) for all \(A\subseteq N_ i\) and \(i\in \{0,...,r(P)- 1\}\) where \(R(A)=\{x\in N_{i+1}\); \(x>a\) for some \(a\in A\}\). Let (P,v) and (Q,w) be weighted posets. A function \(\phi\) : \(P\to Q\) is called a flow morphism, if (i) \(\phi\) is surjective, (ii) \(x<y\) implies \(\phi (x)<\phi (y)\), (iii) \(w(x)=v(\phi^{-1}(x))\) for all \(x\in Q\), (iv) the poset induced on \(\phi^{-1}(x)\cup \phi^{-1}(y)\) is normal and has rank 1 for all x,y\(\in Q\) with \(x<y\). Using the framework of category theory, \textit{L. H. Harper} [Adv. Appl. Math. 1, 158-181 (1980; Zbl 0469.90028)] proved the following theorem: If there is a flow morphism \(\phi\) from (P,v) onto (Q,w), then max \(\{\) v(A); \(A\subseteq P\) is an \(antichain\}=\max \{w(B)\); \(B\subseteq Q\) is an antichain\(\}\). In the paper under review, the author gives another proof of this theorem. This proof is based on the method of linear programming and on the fact that relation graphs of posets are perfect.
0 references
Harper's theorem
0 references
Sperner families
0 references
ranked poset
0 references
weighted posets
0 references
linear programming
0 references
relation graphs of posets
0 references