When do random subsets decompose a finite group?
Let \(G\) be a finite group of size \(n\geq 3\). Let \(a_1,a_2,\dots,a_k,b_1,b_2,\dots,b_k\) be \(2k\) random elements chosen independently from \(G\) and let \(A=\{a_1,\dots,a_k\}\) and \(B=\{b_1,\dots,b_k\}\). The author investigates the event \(\{AB\cup BA=G\}\). In particular the study of this event gives rise to a group invariant \(\theta(G)\) defined as the unique solution between \(1/2\) and 1 of the equation \(2\theta\log|G|=\log\sum_{x\in G}\exp(\theta\log|G|\cdot|C_G(x)|/|G|)\). The author shows that a phase transition occurs as the size of \(A\) and \(B\) passes \(\sqrt{\theta(G)|G|\log|G|}\); i.e. for any \(\varepsilon>0\), if the size of \(A\) and \(B\) is less than \((1-\varepsilon)\sqrt{\theta(G)|G|\log|G|}\), then with high probability \(AB\cup BA\neq G\). If \(A\) and \(B\) are larger than \((1+\varepsilon)\sqrt{\theta(G)|G|\log|G|}\), then \(AB\cup BA=G\) with high probability. The invariant \(\theta(G)\) measures, in some sense, how ``Abelian a group is. For example \(\theta(G)=1\) if and only if \(G\) is Abelian and more in general \(\theta(G)\geq 1/(2-R(G)/|G|)>1/2\), \(R(G)\) being the number of conjugacy classes of \(G\). Moreover if \(G\) is a non-Abelian simple group, then \(\theta(G)=1/2+o(|G|)\).
- A correlation inequality and a poisson limit theorem for nonoverlapping balanced subgraphs of a random graph
- Classical groups, probabilistic methods, and the (2,3)-generation problem
- scientific article; zbMATH DE number 706263 (Why is no real title available?)
- New versions of Suen's correlation inequality
- On conjugacy classes of maximal subgroups of finite simple groups, and a related zeta function.
- The probability of generating a finite classical group
This page was built for publication: When do random subsets decompose a finite group?
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2655780)
