Conjecture on smooth approximations to oscillatory sums

For x1x\geqslant1 and y1y\geqslant1, write Ψ(x,y;t)=nx+(n)ynit\Psi(x,y;t)=\sum_{\substack{n\leqslant x\P^{+}(n)\leqslant y}}n^{-it}.

Smooth approximation conjecture. There exists a constant A>0A>0 such that, uniformly for 1xT1\leqslant x\leqslant T and 2Tt5T2T\leqslant t\leqslant5T,

nx1nit=Ψ(x,y;t)+o(Ψ(x,y)),as T,\sum_{n\leqslant x}\frac{1}{n^{it}}=\Psi(x,y;t)+o\bigl(\Psi(x,y)\bigr),\qquad\text{as }T\to\infty,

where y=(logT+log2x)(log2T)Ay=(\log T+\log^2x)(\log_2T)^A.

The source presents this as the analogue of the Granville–Soundararajan conjecture and notes that it could imply sharpness of the constant in an extreme-value result for ζ(1+it)\zeta(1+it). No resolution is supplied.

Sources & referencesView supporting material

Primary source

Daodao Yang, “Extreme values of derivatives of zeta and L-functions”, arXiv:2204.13826 (2023).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.