On pointwise \(\ell^r\)-sparse domination in a space of homogeneous type (Q2050728)

Let \((S, d, \mu)\) be a space of homogeneous type and let \(X\) and \(Y\) be Banach spaces. For a bounded linear operator \(T\) from \(L^{p_1} (S; X)\) to \(L^{p_1,\infty}(S; Y)\) and \(\alpha \ge 1\) we define the following sharp grand maximal truncation operator \(M^{\#}_{T, \alpha} f (s) := \sup_{B \ni s} \operatorname{ess\,sup}_{s^\prime, s^{\prime\prime} \in B} \|T(f\chi_{S\backslash \alpha B)}(s') - T(f\chi_{S\backslash \alpha B)}(s^{\prime\prime})\|_Y\), \(s \in S\), where the supremum is taken over all balls \(B \subset S\) containing \(s \in S\). The main theorem reads as follows: Let \((S, d, \mu)\) be a space of homogeneous type and let \(X\) and \(Y\) be Banach spaces. Take \(p_1, p_2, r \in [1, \infty)\) and set \(p_0 := \max\{p_1, p_2\}\). Take \(\alpha \ge 3c^2_{d}/\delta\), where \(c_d\) is the quasi-metric constant and \(\delta\) is a constant coming from dyadic system in \((X,d,\mu)\). Assume the following conditions: (i) \(T\) is a bounded linear operator from \(L^{p_1}(S; X)\) to \(L^{p_1,\infty}(S; Y)\). (ii) \(M^{\#}_{T, \alpha}\) is a bounded operator from \(L^{p_2} (S; X)\) to \(L^{p_2,\infty}(S; Y)\). (iii) There is a \(C_r > 0\) such that for disjointly and boundedly supported \(f_1, \ldots, f_n \in L^{p_0} (S; X)\) it holds \(\|T(\sum_{k=1}^n f_k)(s)\|_Y \le C_r (\sum_{k=1}^n\|T f_k(s)\|^r_Y)^{1/r}\), \(s \in S\). Then there is an \(\eta\in (0, 1)\) such that for any boundedly supported \(f\in L^{p_0}(S; X)\) there is an \(\eta\)-sparse collection of cubes in \(S\) such that \(\|T f (s)\|_{Y} \lesssim_{S,\alpha} C_T C_r (\sum_{Q\in S}\langle \|f\|_X\rangle^r_{p_0,Q} \chi_Q(s))^{1/r}\), \(s \in S\), where \(C_T = \|T\|_{L^{p_1}\to L^{p_1,\infty}} + \|M^{\#}_{T, \alpha}\|_{L^{p_2}\to L^{p_2,\infty}}\). \par The author also gives several applications, extends further this result.