Quasirandomness in hypergraphs (Q5915722)

Summary: An \(n\)-vertex graph \(G\) of edge density \(p\) is considered to be \textit{quasirandom} if it shares several important properties with the random graph \(G(n,p)\). A well-known theorem of \textit{F. R. K. Chung} et al. [Combinatorica 9, No. 4, 345--362 (1989; Zbl 0715.05057)] states that many such `typical' properties are asymptotically equivalent and, thus, a graph \(G\) possessing one such property automatically satisfies the others. In recent years, work in this area has focused on uncovering more quasirandom graph properties and on extending the known resulty to other discrete structures. In the context of hypergraphs, however, one may consider several different notions of quasirandomness. A complete description of these notions has been provided recently by \textit{H. Towsner} [Random Struct. Algorithms 50, No. 1, 114--139 (2017; Zbl 1352.05168)], who proved several central equivalences using an analytic framework. We give short and purely combinatorial proofs of the main equivalences in Towsner's result [loc. cit.].