On uniform f-vectors of cutsets in the truncated Boolean lattice

From MaRDI portal
(Redirected from On uniform f-vectors of cutsets in the truncated Boolean lattice)
On uniform \(f\)-vectors of cutsets in the truncated Boolean lattice



Abstract: Let [n]=1,2,ldots,n and let 2[n] be the collection of all subsets of [n] ordered by inclusion. calCsubseteq2[n] is a {em cutset} if it meets every maximal chain in 2[n], and the {em width} of calCsubseteq2[n] is the minimum number of chains in a chain decomposition of calC. Fix 0leqmleqlleqn. What is the smallest value of k such that there exists a cutset that consists only of subsets of sizes between m and l, and such that it contains exactly k subsets of size i for each mleqileql? The answer, which we denote by gn(m,l), gives a lower estimate for the width of a cutset between levels m and l in 2[n]. After using the Kruskal-Katona Theorem to give a general characterization of cutsets in terms of the number and sizes of their elements, we find lower and upper bounds (as well as some exact values) for gn(m,l).


Let \(2^{[n]}\) be the Boolean lattice, i.e. the lattice of subsets of \([n]:=\{1,\dots,n\}\), ordered by inclusion. A family \(\mathcal C\subseteq 2^{[n]}\) is called a cutset if it meets every maximal chain in \(2^{[n]}\). The profile of \(\mathcal C\) is a vector \(f=(f_0,\dots,f_n)\) where \(f_i:=|\{X\in \mathcal C:|X|=i\}|\). Let \(g_n(m,\ell)\) be the smallest value of \(k\) for which the \((n+1)\)-tuple \((f_0,\dots, f_n)\), defined by \(f_i=k\) if \(m\leq i\leq \ell\) and 0 otherwise, can be the profile of a cutset in \(2^{[n]}\). The authors prove: \[ \begin{alignedat}{2} g_n(m,m+1) &=\begin{pmatrix} n-1 \\ m\end{pmatrix} , \quad & & 0\leq m\leq n-1, \\ g_n(m,m+2) & = \sum^m_{j=0} \begin{pmatrix} n-2j-2\\ m-j \end{pmatrix}, \quad & & 0\leq m\leq \frac n2 -1 . \end{alignedat} \] Moreover, if \(\frac nm\) is sufficiently large, \[ \begin{gathered} \begin{pmatrix} n-2\\ m \end{pmatrix} <g_n(m,\ell) \leq \sum^m_{j=0} \begin{pmatrix} n-2j-2\\ m-j \end{pmatrix} , \quad m+2\leq \ell \leq n-m-1, \\ \begin{pmatrix} n-3\\ m \end{pmatrix} <g_n(m,n-m)\leq \sum^m_{j=0} \begin{pmatrix} n-2j-3\\m-j\end{pmatrix}. \end{gathered} \] These numerical results are derived from a qualitativ description of the possible cutset-profiles using compression and canonical forms. The authors conjecture that for sufficiently large \(\frac nm\) \[ g_n(m,\ell) = \begin{pmatrix} n\\ m \end{pmatrix} -\begin{pmatrix} n\\ m-1\end{pmatrix}, \quad 2m\leq \ell\leq n-m-1, \] \[ g_n(m,n-m)=\begin{pmatrix} n-1\\ m\end{pmatrix} -\begin{pmatrix} n-1\\ m-1\end{pmatrix} . \]











This page was built for publication: On uniform \(f\)-vectors of cutsets in the truncated Boolean lattice

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1586331)