Convexity and a Stone-type theorem for convex sets in abelian semigroup setting (Q2346783)

Given a subfield \(\mathbb{F}\) of \(\mathbb{R}\) and a vector space \(X\) over \(\mathbb{F}\), a subset \(A \subseteq X\) is called \(\mathbb{F}\)-convex if for all \(x, y \in A\) and \(t \in [0,1] \cap \mathbb{F}\) the convex combination \(tx+(1-t)y\) belongs to \(A\). Stone's theorem asserts that two disjoint \(\mathbb{F}\)-convex sets can always be separated by two disjoint complementary \(\mathbb{F}\)-convex sets. A completely analogous result was established by the second author [Arch. Math. 52, No. 3, 265--268 (1989; Zbl 0651.20067)] for the separation of disjoint subsemigroups of an abelian semigroup \(S\): given two disjoint subsemigroups \(A_0, B_0 \subseteq S\), there exist two disjoint subsemigroups \(A, B \subseteq S\) such that \(A_0 \subseteq A\), \(B_0 \subseteq B\) and \(A \cup B = S\). In this paper the authors consider two concepts of convexity of subsets of any abelian semigroup. They compare these concepts and examine their algebraic and set-theoretical properties. A Stone-type theorem is presented in this context and two proofs of it are given.