Documentation

FormalConjectures.ErdosProblems.«342»

Erdős Problem 342 #

References:

def Erdos342.UniqueUlamSum (a : ) (n m : ) :

UniqueUlamSum a n m means that $m$ has a unique representation as $a(i) + a(j)$ with $i < j < n$.

Equations
Instances For

    IsUlamSequence a means that $a$ is the Ulam sequence (OEIS A002858): $a(0) = 1$, $a(1) = 2$, and for each $n \geq 2$, $a(n)$ is the least integer greater than $a(n-1)$ that has a unique representation as $a(i) + a(j)$ with $i < j < n$.

    Equations
    • One or more equations did not get rendered due to their size.
    Instances For
      theorem Erdos342.erdos_342.test.a0 (a : ) :
      IsUlamSequence aa 0 = 1

      $a(0) = 1$ by definition.

      theorem Erdos342.erdos_342.test.a1 (a : ) :
      IsUlamSequence aa 1 = 2

      $a(1) = 2$ by definition.

      theorem Erdos342.erdos_342.test.a2 (a : ) :
      IsUlamSequence aa 2 = 3

      $a(2) = 3$: the only pair $(i,j)$ with $i < j < 2$ is $(0,1)$, giving $1 + 2 = 3$.

      theorem Erdos342.erdos_342.test.a3 (a : ) :
      IsUlamSequence aa 3 = 4

      $a(3) = 4$: among sums $> 3$ with a unique representation from $\{1,2,3\}$, the smallest is $4 = 1 + 3$. The candidate $5 = 2 + 3$ is ruled out by minimality since $4$ has a unique representation.

      theorem Erdos342.erdos_342.parts.i :
      True ∀ (a : ), IsUlamSequence a{n : | ∃ (m : ), a m = a n + 2}.Infinite

      Do infinitely many pairs $(a, a+2)$ occur in Ulam's sequence?

      theorem Erdos342.erdos_342.parts.ii :
      True ∀ (a : ), IsUlamSequence ahave d := fun (n : ) => (a (n + 1)) - (a n); p > 0, ∀ᶠ (m : ) in Filter.atTop, d (m + p) = d m

      Does Ulam's sequence eventually have periodic differences? That is, is $a(n+1) - a(n)$ eventually periodic?

      Part (iii), is the density of the sequence 0?