Prove that every subset ##B## of ##A=\{1,...,n\}## is finite

  • Thread starter Thread starter zenterix
  • Start date Start date
  • Tags Tags
    Real analysis
zenterix
Messages
774
Reaction score
84
Homework Statement
Prove that every subset ##B## of ##A=\{1,...,n\}## is finite.
Relevant Equations
Note that the definition of a finite set is that either a set is empty or there exists a bijection between the set and ##\{1,2,...,n\}## for some ##n\in\mathbb{N}##.
I am very unsure about the proof below. I'd like to know if it is correct.

If ##B## is empty then it is finite by definition.

If ##B## is non-empty then since ##B\subset\mathbb{N}## it has a smallest element ##b_1##.

If ##B \backslash \{b_1\}## is non-empty then it has a smallest element ##b_2##.

If ##B \backslash \{b_1, b_2\}## is non-empty then it has a smallest element ##b_3##

If we keep repeating this procedure, we will eventually reach the empty set after a maximum of ##n## steps.

The reason for this is that we obtain an element of ##B## in each step (the smallest element in that step), and if we were to obtain more than ##n## elements then ##B## would have more elements than ##\{1, 2, ..., n\}## of which ##B## is a subset.

Suppose that we reach the empty set after ##m\leq n## steps. We have ##b_1, b_2, ..., b_m## elements obtained from these steps.

Define ##g:B\to\{1,...,m\}## by

##g(b_i)=i## for ##i=1,2,...,m##

Claim: ##g## is a bijection.

Proof

Suppose ##g(b_i)=g(b_j)## for any ##b_i, b_j\in B##.

Then by definition of ##g## we have ##i=j## and hence ##b_i=b_j##

Thus, ##g## is injective.

Now, let ##i\in\{1,...,m\}##. Then ##g(b_i)=i## and so ##g## is surjective.

Thus we have obtained a bijection between ##B## and ##\{1,...,m\}## and so ##B## is finite.
 
Physics news on Phys.org
Using induction might be simpler.
 
For non-empty B it suffices to find a surjection \phi : S\to B for some finite S, since then |B| \leq |S| with equality iff \phi is a bijection.
 
pasmith said:
For non-empty B it suffices to find a surjection \phi : S\to B for some finite S, since then |B| \leq |S| with equality iff \phi is a bijection.
Doesn't finding such a finite ##S## have to go through some procedure like the one in my OP?

That is, don't we have to somehow construct ##S## and isn't that precisely the part of the proof that is the trickiest?

Instead of finding a surjection I found a bijection directly.
 
Induction must be the way to go.
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top