English

On a conjecture of Erd\H{o}s and Graham about the Sylvester's sequence

Number Theory 2025-03-24 v4 Classical Analysis and ODEs

Abstract

Let {un}n=1\{u_n\}_{n=1}^{\infty} be the Sylvester's sequence (sequence A000058 in the OEIS), and let a1<a2< a_1 < a_2 < \cdots be any other positive integer sequence satisfying i=11ai=1 \sum_{i=1}^\infty \frac{1}{a_i} = 1 . In this paper, we solve a conjecture of Erd\H{o}s and Graham, which asks whether lim infnan12n<limnun12n=c0=1.264085. \liminf_{n\to\infty} a_n^{\frac{1}{2^n}} < \lim_{n\to\infty} u_n^{\frac{1}{2^n}} = c_0 = 1.264085\ldots. We prove this conjecture using a constructive approach. Furthermore, assuming that the unproven claim of Erd\H{o}s and Graham that "all rationals have eventually greedy best Egyptian underapproximations" holds, we establish a generalization of this conjecture using a non-constructive approach. [This paper solves Problem 315 on Bloom's website "Erd\H{o}s problems".]

Keywords

Cite

@article{arxiv.2503.12277,
  title  = {On a conjecture of Erd\H{o}s and Graham about the Sylvester's sequence},
  author = {Zheng Li and Quanyu Tang},
  journal= {arXiv preprint arXiv:2503.12277},
  year   = {2025}
}

Comments

23 pages; v2 generalizes the previous results; v3 fixes some typographical errors and adds several remarks; v4 corrects the definition of underapproximation and adds a final section proposing several open problems