Erdős Problem #820
No proof or disproof is known, and it isn't known whether a finite computation could settle it.
What the community database records, not our verdict. erdosproblems.com no longer shows statuses (why).
\(H(n)\) is the smallest integer \(l\) such that there is some \(k < l\) with \((k^n-1, l^n-1) = 1\). We take \(k\) and \(l\) to be integers with \(k \geq 2\), so that \(k^n - 1 \geq 1\) for \(n \geq 1\). Smaller \(k\) give degenerate values: over the integers \(k = 0\) gives \((-1, l^n-1) = 1\), and \(k = 1\) gives \((0, l^n-1) = l^n-1\), which is \(1\) for \(n = 1\), \(l = 2\). Both would contradict the value \(H(1) = 3\) listed in the source. For \(n \geq 1\) the set is nonempty (see `Erdos820.H_spec`). For \(n = 0\) it is empty, a
\(K(n)\) is the smallest integer \(k \geq 2\) such that \((k^n-1, 2^n-1) = 1\). As for `Erdos820.H`, smaller \(k\) are excluded because they give degenerate values. For \(n \geq 2\) the value \(k = 2\) never works, and \(k = 2^n - 1\) always works. For \(n = 1\) we get \(K(1) = 2\), and for \(n = 0\) the set is empty, so `K 0 = 0`.
Attempts
submit one →Read our working report →
The authoritative live page is OPEN but lists “1 claimed proof for this problem”; per the mandatory collision rule, no mathematical attempt was made.
SKIPPED