Erdős Problem a Daythe open tracker

Erdős Problem #360

SOLVEDdatabase updated 2025-08-31

Resolved in some way other than a straight proof or disproof — usual for open-ended questions, or for multi-part problems whose parts went different ways.

What the community database records, not our verdict. erdosproblems.com no longer shows statuses (why).

\(f(n)\): the least number of classes in a partition of \(\{1,\ldots,n-1\}\) such that \(n\) is not a sum of distinct elements of one class. A class contains \(n\) among its subset sums exactly when some finite subset of the class sums to \(n\). For \(n \geq 2\) the partition into singletons is admissible (each element is less than \(n\)), so the set is non-empty and `sInf` is its minimum. For \(n \leq 1\) the set \(\{1,\ldots,n-1\}\) is empty, the empty partition is admissible, and \(f(n) = 0\).

The answer: \(\frac{n^{1/3}(n/\phi(n))}{(\log n)^{1/3}(\log\log n)^{2/3}}\).

Let \(f(n)\) be minimal such that \(\{1,\ldots,n-1\}\) can be partitioned into \(f(n)\) classes so that \(n\) cannot be expressed as a sum of distinct elements from the same class. How fast does \(f(n)\) grow? A problem of Erdős and Graham [ErGr80]. Conlon, Fox, and Pham [CFP21, Theorem 1.5] determined the order of growth of \(f(n)\) up to a multiplicative constant, proving \[f(n) \asymp \frac{n^{1/3}(n/\phi(n))}{(\log n)^{1/3}(\log\log n)^{2/3}}.\] Here \(\phi\) is Euler's totient function, and the statement is asymptotic (\(n \to \infty\)).
Statement as formalized in Formal Conjectures (Google DeepMind, Apache-2.0) · their label: research solved · 2 further variants / known partial results in the Lean file. The authoritative wording is on erdosproblems.com/360.
Prizenone
Topicsnumber theory
Statement in LeanFormal Conjectures
Solution in Leanno
Original pageerdosproblems.com/360
NOBODY HAS SENT US AN ATTEMPT YET

If you take a run at this one — with your own AI subscription or without — the guide says how, and the submission form gets you an honest read of what you found.