Erdős Problem #731 — Find some reasonable function f(n)f(n) such that, for almost all integers nn, the least integer mm such that mmid(2nn)m mid \binom{2n}{n} satisfies m∼f(n).m\sim f(n).

About 51 years old · traced to

Find some reasonable function f(n)f(n) such that, for almost all integers nn, the least integer mm such that mmid(2nn)m mid \binom{2n}{n} satisfies m∼f(n).m\sim f(n).

References

Progress summary

Refreshed
Claimed progress

A new preprint answers the question negatively under a natural smoothness interpretation, while the completely unrestricted version remains open.

Erdős, Graham, Ruzsa, and Straus posed the problem in 1975: determine whether the least integer failing to divide the central binomial coefficient is asymptotic to a reasonable function for almost all inputs.

Known results

  • Erdős, Graham, Ruzsa, and Straus (1975) recorded the coarse estimate A(n)=exp⁡ ⁣((log⁡n)1/2+o(1))A(n)=\exp\!\big((\log n)^{1/2+o(1)}\big) for almost all nn.

June 2026 dyadic-regularity resolution

A preprint proves that no dyadically regular function ff can satisfy A(n)/f(n)→1A(n)/f(n)\to1 in natural density. It also establishes the density-tight scale F(n)=2(log⁡2)1/4(log⁡n)1/4exp⁡ ⁣((log⁡2)log⁡n)F(n)=\sqrt{2}(\log 2)^{1/4}(\log n)^{1/4}\exp\!\big(\sqrt{(\log 2)\log n}\big), while proving persistent multiplicative spread on every sufficiently large dyadic block. The work is formally verified in Lean 44, but its negative conclusion applies only to this explicit formalization of “reasonable.”

Current status (as of June 2026): The problem is resolved negatively for dyadically regular ff, but the original question without that restriction remains open.

Sources

Solutions 0

No solutions have been posted yet.