On the Hurwitz zeta-function with algebraic irrational parameter. II (Q2234366)

Voronin's theorem from 1975 states that the Riemann zeta function is universal in the sense that its shifts approximate a wide class of analytic functions. More precisely, let \(\mathscr{K}\) be the class of compact subsets of the strip \(D=\left\{s \in \mathbb{C}: \frac{1}{2}<\sigma<1\right\}\) having connected complements and let \(H_{0}(K)\) \((K \in \mathscr{K})\), be the class of continuous functions without zeros in \(K\) and analytic inside \(K\). Then Voronin's theorem states that \[ \liminf _{T \rightarrow \infty} \frac{1}{T} \mu\left\{\tau \in[0, T]: \sup _{s \in K}|\zeta(s+i \tau)-f(s)|<\varepsilon\right\}>0 \] for all \(K \in \mathscr{K}, f(s) \in H_{0}(K)\), and \(\varepsilon>0 .\) Here \(\mu\{A\}\) is the Lebesgue measure of a measurable subset \(A \subset \mathbb{R}\). It was later observed that the Hurwitz zeta function \[ \zeta(s, \alpha)=\sum_{m=0}^{\infty} \frac{1}{(m+\alpha)^{s}} \] is also universal if \(\alpha\) is transcendental or rational, different from 1 and \(1/2\). (For algebraic irrational \(\alpha\) the question is still open.) The present paper provides a discrete universality result in the following sense. Suppose that \(\alpha\) is an algebraic irrational number and \(h>0\). Then there exists a nonempty closed subset \(F_{\alpha, h} \subset H(D)\) such that, for every compact set \(K \subset D, f(s) \in F_{\alpha, h}\) and \(\varepsilon>0\), \[ \liminf _{N \rightarrow \infty} \frac{1}{N+1} \#\left\{0 \leq k \leq N: \sup _{s \in K}|\zeta(s+i k h, \alpha)-f(s)|<\varepsilon\right\}>0 \] Moreover, for every compact set \(K \subset D\) and \(f(s) \in F_{\alpha, h}\), the limit \[ \lim _{N \rightarrow \infty} \frac{1}{N+1} \#\left\{0 \leq k \leq N: \sup _{s \in K}|\zeta(s+i k h, \alpha)-f(s)|<\varepsilon\right\}>0 \] exists for all but at most countably many \(\varepsilon>0\). This result is then extended to \(\Phi(\zeta(s,\alpha))\), where \(\Phi\) is a certain operator on \(H(D)\).