Skip to published state

Erdős problem / erdos

no open offer

Problem 535

Exact records and bounded producer offers matched to this problem.

Matching finding records

1 records
  1. vf_95ac77d9b1387ca1

    theoretical

    Erdős Problem #535: declared status 'open'. Formalized: yes. Let r3r\geq 3, and let fr(N)f_r(N) denote the size of the largest subset of {1,,N}\{1,\ldots,N\} such that no subset of size rr has the same pairwise greatest common divisor between all elements. Estimate fr(N)f_r(N). Current best: Erd\H{o}s [Er64] proved the lower boundf3(N)>NcloglogNf_3(N) > N^{\frac{c}{\log\log N}}for some constant c>0c>0, and conjectured this should also be an upper bound. Indeed, the conjectured upper bound would follow from the following stronger version of the sunflower problem: estimate the size of the largest set of integers AA such that ω(n)=k\omega(n)=k for all nAn\in A and there does not exist a1,,arAa_1,\ldots,a_r\in A and an integer dd such that (ai,aj)=d(a_i,a_j)=d for all iji\neq j and (ai/d,d)=1(a_i/d,d)=1 for all ii. The conjectured upper bound for fr(N)f_r(N) would follow if the size of such an AA must be at most crkc_r^k. The original sunflower proof of Erd\H{o}s and Rado gives the upper bound crkk!c_r^kk!. Prize: no. Tags: number theory.

    recordedOpen record