0
$\begingroup$

Let $a(n)$ be the number of positive integers $k$ such that there exists a nonnegative integer $m$ with $k + k^m = n$.

The sequence begins $$0, 1, 1, 2, 1, 3, 1, 2, 1, 3, 1, 3, 1, 2, 1, 2, 1, 3, 1, 3$$ Let $$b(n)=\sum\limits_{i=1}^{\left\lfloor\frac{n-1}{2}\right\rfloor}\left\lfloor\log_{i+1}(n-i)\right\rfloor$$

Also $$c(n)=b(n)-b(n-1)+1$$

I conjecture that $c(n)=a(n+1)$ for $n>0$.

Is there a way to prove it? If the conjecture is true, is it possible to use it to answer the questions posed in A309978? The questions are:

  • Does there exist $n$ such that $a(n) \geqslant 5$?
  • Do there exist examples besides $30$ and $130$ such that $a(n) = 4$?
$\endgroup$
5
  • 2
    $\begingroup$ Did you check the first 1000 terms $a_n$? $\endgroup$
    – markvs
    Commented Apr 3, 2022 at 18:52
  • 1
    $\begingroup$ There is some discussion of this in Dana Mackenzie's paper at gathering4gardner.org/g4g13gift/math/… although the greater concern of that paper is representations as $k^m-k$. Also, oeis.org/A057896 deals with $k^m-k$. $\endgroup$ Commented Apr 3, 2022 at 22:43
  • 3
    $\begingroup$ Not research level. Plus, why the tag "representation theory"? $\endgroup$ Commented Apr 4, 2022 at 5:40
  • 1
    $\begingroup$ I have removed "representation theory" tag. The last 2 questions may be of research level, but I'm not sure. $\endgroup$ Commented Apr 4, 2022 at 14:33
  • 2
    $\begingroup$ For clarification (this confused me for a moment): $a(30) = 4$ because $29 + 29^0 = 15 + 15^1 = 5 + 5^2 = 3 + 3^3$. $\endgroup$ Commented Apr 4, 2022 at 15:40

1 Answer 1

5
$\begingroup$

The formula $c(n)=a(n+1)$ is pretty much straightforward, noticing that $$\lfloor \log_{i+1}(n-i)\rfloor - \lfloor\log_{i+1}(n-1-i)\rfloor=1\quad\text{iff}\quad n-i=(i+1)^m\text{ for some }m.$$ The latter condition means that $n+1=k+k^m$ with $k:=i+1$.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .