Extrapolation of compactness on weighted spaces: bilinear operators (Q2076517)

In a previous paper [``Extrapolation of compactness on weighted spaces'', Preprint, \url{arXiv:2003.01606}] the authors provided the extrapolation of compactness of a linear operator and obtained generalizations of the preceding compact extrapolation to the ``off-diagonal'' and limited range cases . In this paper they extend their results to the bilinear setting. Their main result establishes that if \(\Theta\) is a collection of ordered tripes of Banach spaces and \(T\) is a bilinear operator such that \(T:Y_1\times Y_2\to Y\) is bouunded for all \((Y_1,Y_2,Y)\in \Theta\) and \(T:X_1\times X_2\to X\) is compact for some \((X_1,X_2,X)\in \Theta\) then \(T:Z_1\times Z_2\to Z\) is compact for all \((Z_1,Z_2,Z)\in \Theta\) in several cases of \(\Theta\) involving different cases of weighted \(L^p\) spaces, in particular if \(\vec{s}=(s_1,s_2)\in [1,\infty)^2\) and \(1/s=1/s_1+1/s_2\) and \(\Theta\) consists of all triples \((L^{q_1}(v_1),L^{q_2}(v_2), L^{q}(v_{\vec{v},\vec{q}}))\) where \(\vec{q}=(q_1,q_2)\in (s_1,\infty)\times (s_2,\infty)\), \(1/q=1/q_1+1/q_2<1\), \(v_{\vec{v},\vec{q}}=v_1^{q/q_1}v_2^{q/q_2}\) and \(\vec{v}=(v_1,v_2)\in A_{\vec{q}/\vec{s}}(\mathbb R^{2d})\) or \(\vec{v}=(v_1,v_2)\in A_{q_1/s_1}(\mathbb R^{d})\times A_{q_2/s_2}(\mathbb R^{d})\). The proof uses an abstract tool on interpolation for bilinear maps on Banach function spaces due to [\textit{F. Cobos} et al., Nonlinear Anal., Theory Methods Appl., Ser. A, Theory Methods 199, Article ID 111951, 8 p. (2020; Zbl 1462.46019)]. The main ingredient of the proof consists of getting the different examples of their theorem to fit in the assumptions of the abstract result. As applications they easily recover and improve earlier results on compactness of commutators of bilinear Calderón-Zygmund operators, bilinear fractional integrals and bilinear Fourier multipliers. More general versions of these results have been shown by \textit{M.Cao} et al., [``Extrapolation for multilinear compact operators and applications'', Preprint, \url{arXiv:2011.13191}] but the paper under review avoid the use of some weighted version of Fréchet-Kolmogorov criterion of compactness.