← All problems

Erdős Problem #257 — Irrational reciprocal sums over arbitrary infinite sets

58 years old · traced to

Let N={1,2,3,…}\mathbb N=\{1,2,3,\ldots\}. For an infinite set A⊆NA\subseteq\mathbb N, define

S(A)=∑n∈A12n−1.S(A)=\sum_{n\in A}\frac{1}{2^n-1}.

This series converges, since its positive terms are bounded by those of the convergent series over all n∈Nn\in\mathbb N. Is it true that

S(A)∉QS(A)\notin\mathbb Q

for every infinite A⊆NA\subseteq\mathbb N? Erdős Problem #257

Background

This is a restricted Lambert-series question. Expanding 1/(2n−1)=∑j≥12−nj1/(2^n-1)=\sum_{j\ge1}2^{-nj} and collecting terms by m=njm=nj gives

S(A)=∑m≥1fA(m)2m,fA(m)=#{d∣m:d∈A}.S(A)=\sum_{m\ge1}\frac{f_A(m)}{2^m}, \qquad f_A(m)=\#\{d\mid m:d\in A\}.

For A=NA=\mathbb N, fA(m)=τ(m)f_A(m)=\tau(m), the number of divisors of mm. Erdős proved this full-support instance irrational; it is the base-2 case of the classical Lambert-series connection. Erdős Problem #257 Erdős (1948), On arithmetical properties of Lambert series

References

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.