Hyper-ideals of multilinear operators (Q492758)

An ideal \({\mathcal M}\) of multilinear operators, or a multi-ideal, is a subclass of the class of all multilinear operators between Banach spaces such that, for all \(n \in \mathbb{N}\) and Banach spaces \(E_1, \dots, E_n,F\), each component \({\mathcal M}_n(E_1, \dots, E_n;F) := {\mathcal L}(E_1, \dots, E_n;F)\cap {\mathcal M}\) is a linear subspace of \({\mathcal L}(E_1, \dots, E_n;F)\), contains the \(n\)-linear operators of finite type and if \(A \in {\mathcal M}_n(E_1, \dots, E_n;F)\), \(u_m \in {\mathcal L}(G_m,E_m)\), \(m = 1, \dots, n\), and \(v \in {\mathcal L}(F;H)\), then \(v \circ A \circ (u_1, \dots, u_n) \in {\mathcal M}_n(G_1, \dots, G_n;H)\). This last property means that the ideal is ``stable'' with respect to the composition with linear operators. The authors introduce and develop a more general property, which is called hyper-ideal property, exploring the same stability feature but with respect to the composition with multilinear operators. More than that, they define the concept of hyper-ideal of multilinear mappings, that generalizes the concept of multi-ideals, and establish its general properties comparing with corresponding properties of multi-ideals. Many illustrative examples of hyper-ideals are provided and distinctions with the well-known theory of multi-ideals are presented. It is clear, observing the aforementioned properties, that every hyper-ideal is a multi-ideal. The converse of this statement is false and examples of important multi-ideals that fail to be a hyper-ideal are presented. A feature that also deserves be noted: when a distinguished multi-ideal fails to be a hyper-ideal, the authors construct a hyper-ideal that enjoys the expected property, in the theory of hyper-ideals, of the missing multi-ideal.