1

My understanding of needing a base case, in general, is that after proving the induction step, we can assert that the proposition is true for all values from the base case.

This question ($∀n ∈ Z, n≥12$) $\implies$ ( $∃x, y ∈N$ such that $n= 4x + 5y$) uses 3 bases cases for its strong induction solution:

Base cases: $n = 12, 13, 14, 15$

Clearly, $12= 4(3) + 5(0)$, so $P(12)$ is true.

Also, $13= 4(2) + 5(1)$, so $P(13)$ is true.

And, $14= 4(1) + 5(2)$, so $P(14)$ is true.

And, $15= 4(0) + 5(3)$, so $P(15)$ is true

Meanwhile I only used base case, $n=12$. Their rationale for providing multiple cases is as follows

If we didn’t prove $P(13), P(14), P(15)$ as base cases, then the inductive step to get $k+1 = 13, 14,15$ will fail, since proving these assume that $P(r)$, for $ r < 12$, to be true, which we didn’t prove. By establishing base cases $n = 12$ to $n = 15$, the inductive step can then work forward from $n ≥ 15$.


The first line starting from "If we..will fail" already contradicts with my understanding of the base case that I mentioned above as proving base case of $12$ should suffice to prove inductive steps for $k+1=13,14,15$.

The part where they said "since proving..we didn't prove" doesn't even make sense to me.Why are they talking about $ r < 12$.

And the last sentence says "work forward from n ≥ 15" confuses me, shouldn't we want it to work from $n ≥ 12$.

My main question is essentially why we do need multiple base cases here? The sub-questions are the 3 paragraphs talking about the solution's rationale for providing multiple cases.

Induction Step(included after a comment by fleabloods):

Screenshot of inductive step

Note: If the sub-questions should be separately posted, do let me know

Leon
  • 413
  • FYI, the question being proven using strong induction is one special case of the coin problem. – John Omielan May 30 '20 at 05:28
  • The Frobenius coin problem induction help post also provides similar information. – John Omielan May 30 '20 at 05:36
  • After looking at the links, wouldn't the inductive hypothesis of a strong induction cover those cases? P(base case)∧...P(i) covers the P(base case+3) etc – Leon May 30 '20 at 05:43
  • In normal induction, the inductive step uses the result of the previous value only. However, the answers in those links, and the $2$ answers below, explain that with this particular problem, it's easier when using induction to go in steps of $4$. As such, this is why strong induction in used with $4$ base cases so when your inductive step goes back $4$ values, it guarantees there's a solution. Note the other $3$ base cases don't come from strong induction itself. I don't think I can add much, if anything, more than that to the other answers. Please carefully read & consider them. – John Omielan May 30 '20 at 05:53
  • You didn't include what the induction step actually was..... – fleablood May 30 '20 at 06:48
  • @fleablood Is the induction step a must for the question(since I wanted to ask about the base cases)? – Leon May 30 '20 at 06:51
  • Well, the authors explanation was about why they needed to to the extra base cases because the induction step wouldn't work. We need to see what the induction step was to see why it wouldn't work. – fleablood May 30 '20 at 07:26
  • See... so not the step from 3.2.2. to 3.2.3. The author is assume that because it is true for $12\le i \le k$ then it is true for $k-3$. ... But that assumes $12 \le k-3$. Since $12,13,14 - 3 < 12$ we can't use the induction step yet. We can start using the induction step at $n=15$. It MIGHT have be easier to say: Let our base case be $n=15$. The let $n=12,13,14$ be "sub-base cases". Then if we assume it is true for all cases $k-3 \le i\le k$ we can show it is true for $k+1$. – fleablood May 30 '20 at 15:05
  • I think Omnomnomnom's answer provides a better approach. To paraphrase in my fleablood way: Look instead of doing induction step $P(n)\to P(n+1)$ we do induction step $P(n)\to P(n+4)$. Thus for a base case of $n=i$ we will have proven it for all $n = 4m+i$. But as that isn't all the integers we will have to do it for a base case $n=j$ to get prove it for all $n$ of the form $n=4m+j$. As there are four forms a number can take $4m+0;4m+1;4m+2;4m+3$ we need $4$ base cases. $2$ is our base case for $4m+0$. $13$ is for $4m+1$. $14$ is for $4m+2$ and $15$ for $4m+3$. that covers everything. – fleablood May 30 '20 at 15:13

3 Answers3

1

Note that in the traditional setup for proof for induction, we establish a base case and then prove that $P(k) \implies P(k+1)$. However, it is difficult to come up with a direct proof of this implication for our case. Note in particular that $P(k) \implies P(k+1)$ requires the additional assumption that $k \geq 12$, since the implication fails with $k = 5$ and $k=10$, for instance.

So, instead of proving this implication, we instead use the implication $P(k) \implies P(k+4)$, which does hold in general and is easy to prove.

With this in mind, we can make sense of the statement

If we didn’t prove $P(13), P(14), P(15)$ as base cases, then the inductive step to get $k+1 = 13, 14,15$ will fail, since proving these assume that $P(r)$, for $ r < 12$, to be true, which we didn’t prove.

For example, if we want to use $P(k) \implies P(k+4)$ in order to prove $P(13)$, we would have to first show that $P(9)$ holds. Indeed, if $k=9$, then $P(k) \implies P(k+4)$ tells us that $P(9) \implies P(13)$. This is problematic because we now require $P(r)$ for $r = 9<12$.

I hope that makes everything a bit clearer.

Ben Grossmann
  • 225,327
  • For what it's worth, heres a $P(k)\implies P(k+1)$. If $n=4x + 5y$ then $n+1=4(x-1)+5(y+1)$. But that only works if $x\ge 1$. If $x=0$ then $n = 40 + 5y$ and $n+1 = 44 + 5(y-3)$ But that only works if $y\ge 3$. For $12=43+50$ we can do $n+1=4(x-1)+5(y+1)$ some $3$ times because $x=3$. For $n\ge 15$ we no longer have to worry about if $x \ge 1$ or $y \ge 3$ because one or the other must be true all $n \ge 15$. – fleablood May 30 '20 at 07:44
  • @Omnomnomnom, it seems like your answer is implying we use simple induction with 4 base cases and inductive step of P(k)⟹P(k+4)? Am i on the right track or totally off? – Leon May 30 '20 at 23:50
  • @Leon "Simple induction" would typically refer specifically to using the implication $P(k) \implies P(k+1)$ – Ben Grossmann May 31 '20 at 00:22
1

Your understanding of base case is incorrect. The base cases are simply those cases needed to provide a sufficient foundation for the induction step.

Here the induction step to prove that $P(k+1)$ is true relies on knowing that $P(k-3)$ is true: if there are $x,y\in\Bbb N$ such that $4x+5y=k-3$, then clearly $$4(x+1)+5y=k+1\;,$$ where $x+1,y\in\Bbb N$, so $P(k+1)$ is true. If we’ve verified only that $P(12)$ is true, that argument won’t tell us that $P(13)$ is true: in that case $k=12$, so $k-3=9$, and we’d need to know that $P(9)$ is true in order to use that argument. But as your source says, we didn’t verify that $P(9)$ is true, so that argument simply isn’t valid.

As it happens, $P(9)$ is true, so if we’d checked it as a base case, we could have used the induction step to see that $P(13)$ is true. $P(10)$ is also true, so we could have verified it as part of the base case and used the induction step to see that $P(14)$ is true. But $P(11)$ is not true, so there is no possible way to use the induction step to derive $P(15)$. It turns out that $11$ is the last failure: $12$ is the smallest integer such that $P(n)$ is true for it and every larger integer.

The induction works only after we know that $P(n)$ is true for four consecutive integers $n$: if $P(n)$, $P(n+1)$, $P(n+2)$, and $P(n+3)$ are true, the truth of $P(n)$ allows us to use the induction step to prove that $P(n+4)$ is true, the truth of $P(n+1)$ allows us to use the induction step to prove that $P(n+5)$ is true, and so on. But we need those four consecutive integers: if $P(n)$ and $P(n+1)$ were true, but $P(n+2)$ were not, the induction step would give us the truth of $P(n+4)$ and $P(n+5)$, but not the truth of $P(n+6)$, because for that we’d need to know that $P(n+2)$ was true.

Thus, the induction argument in this example really does use all four parts of the base case, $P(12)$, $P(13)$, $P(14)$, and $P(15)$: without them, the argument falls apart at one of the next four cases.

Brian M. Scott
  • 616,228
1

You didn't include what the induction step was!

We can't evaluate their argument if we don't know what the induction step was.

But from their commentary I suspect the induction step is something like:

If it is true for $n$ and for all values between $12$ and $n$ then it should be true for $n-3$ (assuming $n-3$ is between $12$ and $n$); then $n-3 = 4x + 5y$ for some $x,y$. So $n+1 = 4(x+1) + 5$.

But what's wrong with that?

Let's try it on $12$. Let's try to go from $12$ to $13$. It is true for $12$ and for all values between $12$ and $12$. So it should be true for $12-3=9$ (assuming $12-3=9$ is between $12$ and $12$) and ..... ....!!!!skreech!!!!!.....

NO! $9$ is not between $12$ and $12$ so we can't use the induction step.

Okay we'll can figure out a special case for $13$. So we now have that is it true for all $13$ and all numbers between $12$ and $13$.

Let's try to use the induction step to go from $13$ to $14$. It is true for $13$ and for all values between $12$ and $13$. So it ought to be true for $13-3=10$ so long as $10$ is between.... Oh G#@%!

So we do a special case for $14$.

But, lord, how long can this go on?! How many special cases will we have to do before we can do the induction step and never worry about special cases?

Well, we can do our induction step if we can assume that $n-3$ is between $12$ and $n$. But for $n-3$ to be between $12$ and $n$ we must have $n$ at least $15$. So we can't use our induction step untill we have gotten to $15$

So we need special base cases for $12,13,14$ and $15$.

======

Note: We do have $12 -3 =9$ and we can do $9 = 1*4 + 1*5$ so $13 = (1+1)*4 + 1*5 = 2*4 + 5$. But we have to have shown $9$ was possible first.

Likewis we could have gotten $14$ from $10$. But we can't have gotten $15$ from $11$ because $11 = 4x + 5y$ is impossible. Note: $15 = 0*4 + 3*5$ is the only way to do this and there are not $4$s so we can't have gotten $15$ from $11$ but just adding only more $4$.

fleablood
  • 124,253
  • I've been re-reading the answer a few times(I lied, I have been reading them for hours). But after I read your comment on including the induction step, something kicked in. So if I'm following correctly, the induction step uses P(k-3), we obviously want k-3≥12 so k≥15. Since our induction step only accounts for k≥15, we need to include 12,13,14,15 as base cases. – Leon May 30 '20 at 09:51
  • Just to test if I understood correctly, we can use P(k-4) as well right with the additional base case of 16? Or P(k-7) with ton more base cases? – Leon May 30 '20 at 09:55
  • I have a feeling that what I did was flawed – Leon May 30 '20 at 10:31
  • 1
    To be honest. If I surmise the induction step Is $P(n)\to P(n+1)$ via $P(n-3)\implies P(n-3+4)=P(n+1)$ I think the author is being a bit misleading and obtuse in trying to hammer and induction step into the familiar $P(n)$ to $P(n+1)$ format. Then actual $P(n)$ has nothingto do with concluding $P(n+1)$. Instead this is a $P(n)\implies P(n+4)$ induction step. Which is fine but need $4$ base case as induction on base case $i$ prove $P(4m+i)$ and we have $4$ equvialencies of $n = 4m+i$ – fleablood May 30 '20 at 14:56
  • @fleablood: It’s especially silly to try to shoehorn it into that mold when it’s apparently being explicitly identified as an instance of strong induction, for which the default induction hypothesis when proving $P(k)$ ought to be that $P(n)$ holds for all $n$ such that $n_0\le n<k$. After the fact one can point out that in this case it’s really $P(k-4)$ that’s crucial and go on to explain why that necessitates establishing it for four consecutive numbers. – Brian M. Scott May 30 '20 at 15:32
  • @BrianM.Scott Yeah. That's what bugs me. $P(n)$ is never used to show $P(n+1)$. The only thing used is $P(n-3)$ and $P(n)$ is never used to show $P(n-3)$ except in the passive sense $12 \le n-3 \le n$ and we know that $P(k)$ is true for all $12\le k\le n$. And then just to be really irritating, we don't know that $12 \le n-3 \le n$. – fleablood May 30 '20 at 16:22