Degree asymptotics of the numerical semigroup tree. (Q2447250): Difference between revisions

A numerical semigroup \(S\) is an additively closed submonoid of the natural numbers with finite complement. The largest element of this complement is called the Frobenius number of \(S\) and the size of the complement is called the genus of \(S\). Many recent papers have investigated enumerative questions about semigroups. In particular, \textit{M. Bras-Amorós} conjectured [in Semigroup Forum 76, No. 2, 379-384 (2008; Zbl 1142.20039)] that the number of numerical semigroups of genus \(g\) is asymptotic to a constant times \(\varphi^g\) where \(\varphi\) denotes the golden ratio. This conjecture has been proven by \textit{A. Zhai} [Semigroup Forum 86, No. 3, 634-662 (2013; Zbl 1276.20066)] by carefully analyzing the semigroup tree. Every numerical semigroup has a unique minimal generating set, and removing a generator larger than the Frobenius number gives another semigroup with genus one larger. This process generates a tree where the nodes at depth \(g\) correspond to semigroups of genus \(g\). The author builds on Zhai's work to prove a conjecture of Lynnelle Ye, that the proportion of semigroups of genus \(g\) with exactly \(h\) children in the semigroup tree approaches \(\varphi^{-(h+2)}\) as \(g\) goes to infinity. A main idea is to approximate the number of semigroups of genus \(g\) and \(h\) children by the number of these that are `strongly descended', those for which the Frobenius number plus the smallest nonzero element of the semigroup is a minimal generator. The combinatorial analysis in this paper is quite intricate but the author does a great job explaining the main ideas and making the arguments easy to follow.