Weighted boundedness of multilinear maximal function using Dirac deltas (Q2174833)

Given two families \(\{K^1_i\}_{i \in \mathbb N}\), \(\{K^2_i\}_{i \in \mathbb N}\) of locally integrable kernels on \(\mathbb R^n\), the authors consider the bilinear operators \[ T_i(f_1, f_2)(x):=(K^1_i\ast f_1)(x)(K^2_i\ast f_2)(x), \quad i \in \mathbb N, \ \ x \in \mathbb R^n, \] and the bilinear maximal operator \[ T^*(f_1, f_2)(x):=\sup_{i \in \mathbb N} |T_i(f_1, f_2)(x)|, \quad x \in \mathbb R^n. \] Then, under certain conditions, they show that the weighted weak-type boundedness of the maximal operator \(T^\ast\) is equivalent to the weighted weak-type boundedness of \(T^\ast\) on the finite linear combinations of Dirac deltas, i.e., if \(q>0\), then the operator \[ T^\ast: L^1(w_1) \times L^1(w_2)\rightarrow L^{q,\infty}(v) \] is bounded if and only if \[ v\left\{x \in \mathbb R^n; \,T^\ast\left ( \sum_{i=1}^H \delta_{a_i},\sum_{j=1}^G \delta_{b_j}\right)(x)>\lambda\right\}<\frac{C}{\lambda^q}\left ( \sum_{i=1}^H w_1(a_i)\right)^q \left ( \sum_{j=1}^G w_2(b_j)\right)^q \] for all \(\lambda >0\) and \(H, G \in \mathbb N\), where \(a^\prime_i\)s and \(b^\prime_j\)s are distinct points in \(\mathbb R^n\). The result is applied to the bilinear Hardy-Littlewood maximal function. The multilinear case can be treated analogously.