← All problems

Erdős Problem #358 — Consecutive subsums with unbounded representation counts

49 years old · traced to

Let A=(ai)i≥1A=(a_i)_{i\geq 1} be a strictly increasing infinite sequence of integers. For each integer nn, let f(n)f(n) be the number of pairs of positive integers (u,v)(u,v) with u≤vu\leq v for which the sum over the consecutive block from uu through vv equals nn:

f(n)=#{(u,v):1≤u≤v, ∑i=uvai=n}.f(n)=\#\left\{(u,v):1\leq u\leq v,\ \sum_{i=u}^{v}a_i=n\right\}.

Does there exist such a sequence for which f(n)→∞f(n)\to\infty as n→∞n\to\infty? The question also asks whether one can have f(n)≥2f(n)\geq 2 for every sufficiently large nn; this is a weaker target than divergence of f(n)f(n). Erdős Problems #358

Background

The problem is recorded in Erdős’s 1977 discussion of combinatorial number theory and in the 1980 collection by Erdős and Graham. A simple benchmark is ai=ia_i=i: for positive nn, the number of consecutive-block representations is the number of odd divisors of nn. The 1980 collection also asks whether every sufficiently large integer can be represented at least once. That condition alone is immediate for ai=ia_i=i, since singleton blocks represent every positive integer; requiring at least two representations avoids this triviality. Erdős Problems #358 Erdős (1977), Problems and results on combinatorial number theory. III Erdős and Graham (1980), Old and new problems and results in combinatorial number theory

A related question studied by Moser and Erdős takes AA to be the primes. They conjectured that the limsup of the number of representations is infinite, but could not prove even that the integers admitting such a representation have positive upper density. In modern additive-combinatorial language, the consecutive-block count can be viewed as a self-correlation of the indicator of a convex set of integers. Erdős Problems #358 Moser (1963), Notes on number theory. III. On the sum of consecutive primes

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.