Let G be a graph with V(G)={1,...,2n} E(G)={{2i-1,2i}: 1<=i<=n} In general how many automorphisms (including the identity automorphism) are there for G?
You're going to have to use the definition of automorphism here.
I'm a little unsure of exactly what an automorphism implies. Does it need to retain the same shape as the original graph?
Going to need a proper definition.
for two graphs to be isomorphic there must be a bijection f:V(G1)->V(G2) such that verticies f(u)and f(v) are adjacent in G2 iff u and v are adjacent in G1. An Isomorphism from G to itself is called an automorphism.
this page shows an example of isomorphism http://mathworld.wolfram.com/GraphAutomorphism.html
Well, if we considered n=2, then the vertices are {1,2,3,4} The edges would be {(1,2), (2,4)}
I got 16 automorphisms when I tried n=2. But I'm not sure if all of them are correct
If you changed 2 to be any other number, then you must assign 1 and 4. Other than that, there are no restrictions.
How did you get 16?
I don't know if this is the correct way of showing an automorphism
What you did doesn't make sense. Suppose your function would do f(1) = 2. Then we have to let f(2)=1.
I'm not sure what nodes you are swapping with what...
An automorphism is a function which maps each node to a new node, but maintains adjacency, roughly speaking. Correct?
yeah
If there were no edges, then we could say the number of automophisms is \(|V|^2\), and this is the maximum number since we don't have any extra restrictions when we don't have any edges.
Whoops, that isn't actually correct either...
The maximum automorphisms, which no edge restrictons, would just be \(|V|!\) since we're just permuting vertices...
Does that make sense?
I think so
So we know it has to be between \(1\) and \(4!=24\)
When n=2 V={1,2,3,4} E={(1,2),(3,4)} We can let 1 be 1 or 2. But 2 must then be the other.
So for 1, we can pick 4 ways to pick it. This decision forces us to pick a certain node for 2, basically it means there will only be one way to pick 2. For 3, we have 2 remaining nodes to pick. for 4, just like 2, we have only one way to pick it.
Hmm, 4*1*2*1=8
that makes sense to me, but I'm still having a little trouble with the definition itself. Does it need to maintain the shape of the original graph? For example can I create cross edges like I did?
If I keep the same shape I get 8 automorphisms when n=2
Well, the best way to write an automorphism is to list the vertices, then write the vertice they are mapped to next to them.
ok this is getting more clear
so write a table ``` V | f(V) 1 | 2 2 | 1 3 | 3 4 | 4 ```
ok I think this makes sense
The next step is to list each edge as \((v_1,v_2)\) and making sure that \((f(v_1),f(v_2))\) is still in the list.
For example ``` E | f(E) (1,2) | (2,1) (3,4) | (3,4) ```
I think all the graphs this creates are just more and more lines with 2 vertices. There will be n of them I think. The first line can map in 2 ways to n lines. The next can map in 2 ways to n-1 lines and so on. So maybe there are 2*n! ways?
Just a thought I had
But 2*2! = 4, but we want 8 when n=2.
hmm you're right. That's not quite right.
I think I have the answer already.
Is my thinking close at all?
or is there another way of going about this?
Your thinking is a start, but it isn't really close to an answer.
The way you should be thinking about this is as follows... We have 2n nodes. How many ways do we assign node 1? How many ways do we assign node 2? etc.
there are 2n ways to assign node 1 there is 1 way to assign node 2 there are 2n-1 ways to assign node 3 there is one way to assign node 4 so half the choices are one choice. So we want a factorial type thing on 2n where it goes down by 2 each time
There isn't 2n-1 ways to assign node 3, since 1 and 2 were already assigned.
oh right 2n-2 ways of assigning node 3
We have: \[ (2n)(2n-2)(2n-4)...(2) = (2\times n)(2\times (n-1))...(2\times 1) \]
(2n)(2n-2)(2n-4)(2n-6)... 2(n)(n-1)(n-2)(n-3)... looks like 2(n!) again
\[ =\underbrace{2\times 2\times \ldots \times 2}_{n}\times n! = 2^nn! \]
oooh
I made a mistake factoring out the 2
It's kinda lame, but it turns out that \(2^nn! = n!!\) I say lame because \(n!!\neq (n!)!\) due to this definition.
Whoops, I said that wrong... \[ n!! = n(n-2)(n-4)\dots (2) \]So when \(n\) is even, we can say: \[ n!! = 2^{n/2}(n/2)! \]
So basically \(2^nn! = (2n)!!\). Regardless, the answer is probably best left at \(2^nn!\)
You could also write it as: \[ \prod_{i=1}^{n}2i \]
\[ \prod_{i=1}^{n}2i = \left(\prod_{i=1}^{n}2\right)\left(\prod_{i=1}^{n}i \right) = (2^n)(n!)=2^nn! \]
ok thank you very very much for helping me on this
I was pretty stuck on it
would you mind possibly helping my on just one more?
I suppose.
Let G be a graph on 5 verticies. Show that it is not possible for every vertex to have degree 3
where the degree is the number of edges incident with a vertex
when I try to do it I can't make it work. I just don't really know why it's happening
oh I think I've got it!
ok
I have a theorem that states\[\sum_{v \in( G )}^{} \deg(v) = 2|E(G)|\] since each edge has 2 ends and when we sum the degrees we are counting each edge twice once for each edge. Given that this is true. If we assume that the degree of each vertex is 3 then the sum is 15. This gives us that 15 =2|E(G)| which is a contradiction because 15 is not a multiple of 2.
That makes sense.
woo thanks so much for helping me! I really really appreciate it!
Join our real-time social learning platform and learn together with your friends!