number-theory / Number Theory, Additive Bases

Erdős Problem #326

Does there exist $A=\{a_1<a_2<\cdots\}\subset \mathbb{N}$ which is a minimal basis of order $2$ (every large integer is the sum of $2$ elements from $A$, and no proper subset of $A$ has this property) such that $\lim_{k\to \infty}a_k/k^2=c$ for some $c\neq 0$? A claimed construction gives a minimal basis with $A(x)=C\sqrt{x}+O(1)$, answering the question affirmatively; Erdős and Graham had conjectured a negative answer.

10Significance / 100
1Frontier events
0Verification tasks
0Recorded attempts

Temporal state

Current frontier

No reconciled state yet.

Append-only history

Frontier timeline

number-theoryJun 14, 2026Significance 10/100Registry: lean verified

Erdős Problem #326

Prior state unknownproved

Affirmative answer claimed, contrary to the negative answer Erdős and Graham conjectured; erdosproblems.com still lists the problem open

SourceReplayReproducedFormal proofStatement auditExternal checkExpert reviewPeer review

Research memory

Claims and attempts

Scoped claims

Source authenticated

Does there exist $A=\{a_1<a_2<\cdots\}\subset \mathbb{N}$ which is a minimal basis of order $2$ (every large integer is the sum of $2$ elements from $A$, and no proper subset of $A$ has this property) such that $\lim_{k\to \infty}a_k/k^2=c$ for some $c\neq 0$? A claimed construction gives a minimal basis with $A(x)=C\sqrt{x}+O(1)$, answering the question affirmatively; Erdős and Graham had conjectured a negative answer.

Affirmative answer claimed, contrary to the negative answer Erdős and Graham conjectured; erdosproblems.com still lists the problem open

Recorded attempts

Evidence graph

Connected research record

No public relationships recorded yet.