1

Since the last post no one gave the solution, so i reopen one and use other approach searched in this forumn. https://math.stackexchange.com/a/91323/620871

Show that the polynomial function $$f(m,n)=(m+n−2)(m+n−1)/2+m $$ is one-to-one and onto. Both domain is $\Bbb Z^+\times \Bbb Z^+$, codomain are $\Bbb Z^+$.

I want to prove $f(m,n)=f(p,q) \longrightarrow (m=p \text{ and }n=q)$.

$$\frac12(m+n-2)(m+n-1)+m=\frac12(p+q-2)(p+q-1)+p\;.\tag{1}$$ The first step is to show that $m+n=p+q$, so suppose not. We may as well assume that $m+n<p+q$. For convenience let $a=m+n$ and $d=(p+q)-a$, so that it becomes $$\frac{(a-2)(a-1)}2+m=\frac{(a+d-2)(a+d-1)}2+p\;.$$

Then $$\begin{align*} m-p&=\frac{(a+d-2)(a+d-1)}2-\frac{(a-2)(a-1)}2\\ &=ad+\frac{d(d-3)}2\\ & \end{align*}$$

Since $a\ge 2$,$d\ge1$, we dicovered that $m-p>a$ when $a\ge 2$,$d>1$,$m-p>m+n,$which is absurd.

However,if $d=1$,$a\ge 2$,then $m-p>a$ is not hold. How to duel with the case that $d=1$,$a\ge 2$ or we can just ignore it?

Thanks.

Moly Holy
  • 115

1 Answers1

1

There’s at least one problem with what you’ve done. You assume that $m+n>p+q$, let $a=m+n$ and $d=p+q-a$; clearly this implies that $d<0$, so the later assertion that $d\ge 1$ cannot be right. I suggest a slightly different approach.

We know that

$$\frac{(m+n-2)(m+n-1)}2=\sum_{k=1}^{m+n-2}k$$

and

$$\frac{(p+q-2)(p+q-1)}2=\sum_{k=1}^{p+q-2}k\;.$$

Without loss of generality assume that $m+n\ge p+q$. Then

$$0=f(m,n)-f(p,q)=\sum_{k=p+q-1}^{m+n-2}k+m-p\;.\tag{1}$$

(If $p+q-1>m+n-2$, the summation evaluates to $0$.) If $m+n>p+q$, then $m+n-2\ge p+q-1$, and $(1)$ implies that

$$0=\sum_{k=p+q-1}^{m+n-2}k+m-p\ge p+q-1+m-p=m+q-1\ge 1\;,$$

which is absurd, so $m+n=p+q$, and $(1)$ implies that $0=m-p$, i.e., that $m=p$, which in turn implies that $n=q$. Thus, $f$ is one-to-one.

When you try to prove that $f$ maps $\Bbb Z^+\times\Bbb Z^+$ onto $\Bbb Z^+$, you may find the following diagram helpful; each point $\langle m,n\rangle$ is labelled with the number $f(m,n)$.

$$\begin{array}{ccc} n&\begin{array}{c|cc} 4&7&12&18&25\\ 3&4&8&13&19\\ 2&2&5&9&14\\ 1&1&3&6&10\\\hline &1&2&3&4 \end{array}\\ &m \end{array}$$

Notice that the points are numbered along consecutive diagonals from upper left to lower right, and that each diagonal has one more point on it than the previous one.

Brian M. Scott
  • 616,228
  • Thanks,the alternative appraoch is good.But i have a typo what if p+q>m+n?How to duel with the case that d=1,a≥2 or we can just ignore it? – Moly Holy Jun 19 '20 at 04:40