The strongest intrinsic meaning of sequential-evaluation convergence (Q870248)

Let \(X\) be a Banach space. For a Fréchet space \(E\) and the vector sequence space \(\lambda (X) \subset X^{\mathbb N}\), where \( \lambda (X) \in \{ \ell _{\infty }(X),c_{0}(X),\ell _{p}(X)\) \((p>0) \} \), the authors consider the \(\beta\)-dual \(\lambda (X)^{\beta E}:=\{(A_{j}) \subset E^{X}\mid \forall (x_{j}) \in \lambda (X) :\sum_{j}A_{j}(x_{j})\) converges\} (here \(E^{X}\) is the vector space of all (not necessarily linear) mappings \(A:X\to E\)). They characterize subsets \( M\subset \lambda (X) \) in which the series \(\sum_{j}A_{j}(x_{j})\) converges uniformly for each \((A_{j}) \in \lambda (X)^{\beta E}\). For example, Theorem 2.2 asserts: For \(M\subset \ell _{p}(X)\), where \(p>0\), the following conditions are equivalent: (1) \(\lim_{n}\sum_{j\geq n} \| x_{j}\| ^{p}=0\) uniformly for \((x_{j}) \in M\); (2) for every Fréchet space \(E\) and \((A_{j}) \in \lambda (X)^{\beta E}\), the series \( \sum_{j}A_{j}(x_{j}) \) converges uniformly in \(M\). The final part of the paper is devoted to pointwise convergence of sequences \((A^{(n)}) \subset [\lambda(X)]^{\beta Y}:= \{(A_{j})\in \lambda (X)^{\beta Y}\mid \forall j\in \mathbb{N}:A_{j}(0) =0\}\), where \(Y\) is a Banach space. For example, it is proved that \(\lim_{n}\sum_{j}A_{j}^{(n) }(x_{j}) =\sum_{j}A_{j}(x_{j})\) for each \((x_{j}) \in \lambda (X)\) if and only if (a) \(\lim_{n}A_{j}^{(n) }(x) =A_{j}(x)\) for each \(x\in X\) and \(j\in \mathbb N\), and (b) \( \sum_{j}A_{j}^{(n) }(x_{j})\) converges uniformly for both \(n\in \mathbb N\) and \((x_{j})\) in any precompact subset of \(\lambda (X)\).