Pietsch-Maurey-Rosenthal factorization of summing multilinear operators (Q1624177)

Analyzing fundamental ideas of Pietsch, Rosenthal, and Maurey, it was proved in [\textit{A. Defant}, Positivity 5, No. 2, 153--175 (2001; Zbl 0994.47036)] that, for every $q$-concave bounded (linear) operator $T: X \to E$, where $X$ is a $q$-convex order continuous Banach lattice over the measure space $(\Omega, \Sigma, \mu)$ and $E$ a Banach space, there is a bounded multiplication operator $M_g: X \to L_q(\mu)$, $f \mapsto fg$, and a bounded operator $S: L_q(\mu) \to E$ such that $T = S M_g$. This is a straightforward consequence of what nowadays is called a domination theorem -- the prototype being Pietsch's domination theorem for $q$-summing operators. In [\textit{O. Delgado} and \textit{E. A. Sánchez Pérez}, Positivity 20, No. 4, 999--1014 (2016; Zbl 1436.46032)], the notion of $p$-strongly $q$-concave operators was invented in order to show that a reasonable variant of the preceding factorization theorem holds true although the assumptions on $T$ are considerably weakened. As its main purpose, the article under review establishes the new setting of $(p_1, \ldots, p_m)$-strongly $(q_1, \ldots, q_m)$-concave multilinear operators $T$ on products $X_1 \times \ldots \times X_m$ of Banach lattices which allows to transfer results so far known for linear operators to multilinear ones. Multilinear domination and factorization theorems are proved (in and outside of the framework of Fremlin tensor products). The results also apply to classes of linear operators defined by certain vector-valued multilinear norm inequalities -- an example being the classical Banach operator ideal of $q$-dominated operators.