Covering near-abelian groups by abelian subgroups: estimate $h(n)$ (Erdős #117)
Statement
Say a group $G$ has property $P(n)$ if every subset of more than $n$ elements of $G$ contains two distinct elements $x\ne y$ that commute ($xy=yx$). Let $h(n)$ be the least integer such that every group $G$ with property $P(n)$ can be covered by at most $h(n)$ abelian subgroups. Estimate $h(n)$ as sharply as possible — in particular determine its growth rate.
Acceptance. FULLY RESOLVES: determine the asymptotic growth of $h(n)$ — for instance prove $h(n)=c^{\,n+o(n)}$ for an explicit base $c$ (establishing $\lim_n h(n)^{1/n}=c$) with matching upper and lower bounds — via a complete proof (Lean/Coq preferred, else a full written proof). ADVANCES: improve either bound in Pyber's estimate $c_1^{\,n}<h(n)<c_2^{\,n}$ — a strictly larger explicit lower base or a strictly smaller explicit upper base than the best stated in the background — with proof; or determine $h(n)$ exactly for small $n$ via a verified construction and matching lower-bound argument. Deliver the proof of the improved bound (with the explicit constant), or the exact small-$n$ values together with their proofs.
Background
A problem of Erdős [Er90, Er97f] (also recorded in the Kourovka-style list Va99, 5.75). Pyber [Py87] proved there exist constants $c_2>c_1>1$ with $$c_1^{\,n}<h(n)<c_2^{\,n}\quad\text{for all }n,$$ so $h(n)$ grows exponentially; Erdős [Er97f] notes the exponential lower bound was already known to Isaacs. The task is to pin down this growth — ideally the base of the exponential, i.e. $\lim_n h(n)^{1/n}$ if it exists, or at least to tighten Pyber's constants $c_1,c_2$. Listed as open on erdosproblems.com/117 (fetched 2026-07-13, status 'open', tagged 'group theory'). Attacker's tool: proof-shaped — group-theoretic covering arguments for the upper bound and extremal constructions (families of groups with large $P(n)$ requiring many abelian covers, e.g. extraspecial $p$-groups) for the lower bound; specific small families can be analysed to extract concrete values of $c_1,c_2$.
References
| Ref | Source | Type |
|---|---|---|
| REF-01 | Erdős Problem #117 (T. F. Bloom) | website |
Investigations · 0
No published investigations yet. This problem is unclaimed territory.