Geometry and topology of random 2-complexes (Q891189)

A model producing random simplicial complexes was recently suggested and studied in [\textit{N. Linial} and \textit{R. Meshulam}, Combinatorica 26, No. 4, 475--487 (2006; Zbl 1121.55013)] and [\textit{R. Meshulam} and \textit{N. Wallach}, Random Struct. Algorithms 34, No. 3, 408--417 (2009; Zbl 1177.55011)]. The model from the first cited article generates a random \(d\)-dimensional complex \(Y\) by considering the full \(d\)-dimensional skeleton of the simplex on the vertices \(\{1,..., n\}\) and retaining \(d\)-dimensional faces independently with probability \(p\). The results of both articles provide threshold functions for the vanishing of the \((d-1)\)-st homology groups of random complexes with coefficients in a finite abelian group. Significant progress in understanding the topology of random 2-complexes was made in [\textit{E. Babson, C. Hoffman} and \textit{M. Kahle}, J. Am. Math. Soc. 24, No. 1, 1--28 (2011; Zbl 1270.20042)] where the authors investigated the fundamental group of random 2-complexes and showed that the fundamental group of a random 2-complex is either nontrivial and Gromov hyperbolic (for \(p\ll n^{-1/2 -\varepsilon}\)) or trivial (if \(p^2 n - 3 \log n\to \infty\)) asymptotically almost surely. In [\textit{D. Cohen} et al., Discrete Comput. Geom. 47, No. 1, 117--149 (2012; Zbl 1237.55009)] it was proved that a random 2-complex \(Y\) collapses to a graph if \(p\ll n^{-1}\) and therefore its fundamental group is free. The paper [\textit{L. Aronshtam} et al., Discrete Comput. Geom. 49, No. 2, 317--334 (2013; Zbl 1260.05171)] suggests an explicit constant \(\gamma\) such that for any \(c<\gamma\) a random \(Y \in Y(n, c/n)\) is either collapsible to a graph or contains a tetrahedron. It is important to have a model producing random aspherical 2-dimensional complexes \(Y\). A connected simplicial complex \(Y\) is said to be aspherical if \(\pi_i(Y)=0\) for all \(i\geq 2\). For 2-dimensional complexes \(Y\), asphericity is equivalent to the vanishing of the second homotopy group \(\pi_2(Y)=0\). Random aspherical 2-complexes could be helpful for testing probabilistically some open problems of two-dimensional topology, such as the Whitehead conjecture. This conjecture claims that a subcomplex of an aspherical 2-complex is also aspherical. A survey of results related to the Whitehead conjecture can be found in [\textit{S. Rosebrock}, Sib. Èlektron. Mat. Izv. 4, 440--449 (2007; Zbl 1299.57005)]. In the present paper the authors revisit the hyperbolicity theorem from [Babson, Hoffman, Kahle, loc. cit.] having in mind two principal goals: firstly, the authors simplify and make more transparent the main ideas of the proof, secondly they observe that the arguments there give in fact a slightly stronger statement: a uniform lower bound on the isoperimetric constants of all subcomplexes of random complexes.