(Graph Theory) Adjacency Matrix

This post will briefly go over the definition of adjacency matrix for graphs, some related theorems regarding graph isomorphism, and some proofs.

Adjacency Matrix

• The adjacency matrix is one way to formally represent a graph, for example in convenience for computer computation.
• Given a graph $G:=(V,E)$, where $|V|=n$, to obtain the adjacency matrix of this graph, we firstly select an ordering of the vertices, say we (re)name them as $v_1,...,v_n$.
• Next, we label the rows and columns of a matrix with the ordered vertices.
• The entry in this matrix in row $i4, column $j$, when $i\ne j$, is the number of edges incident on $i$ and $j$.
• If $i=j$, the entry is twice the number of loops incident on $i$, and this is because we want the nice formula ${\sum }_{i=1}^n\delta (v_i)=2|E|,n\ge 1$ to hold in general ($n=0$ is not particuarly interesting).

Remark

• Besides entries on the main diagonal, the information are redundantly stored twice in the adjacency matrix (the information above the main diagonal is identical to the information below the main diagonal). Therefore, to obtain $\delta (v_i)$, we sum up all the entries in the $i$th row OR all the entries in the $i$th column, but never both.

Reordering the Vertices

• Ordering affects which row represents which vertex. Observe that if a graph has $n$ vertices, we have $n!$ ways to reorder the vertices, and we can thus produce $n!$ (not necessarily distinct) adjacency matrices.

Graph Isomorphism and Adjacency Matrix

Proposition

• Two graphs $G_1,G_2$ are isomorphic iff there is some ordering of their vertices so that $A_1=A_2$ ($A_1$ is an adjacency matrix of $G_1$, and $A_2$ is an adjacency matrix of $G_2$).

Proof

Isomorphism$⟹$Ordering

• Suppose $G_1,G_2$ are isomorphic, then there exist bijections $f:V_1\to V_2$ and $g:E_1\to E_2$ such that $e\in E_1$ is incident on $v,w\in V_1$ iff $g(e)\in E_2$ is incident on $f(v),f(w)\in V_2$.
• Thus, we can choose arbitrarily an ordering of the vertices of $G_1$, say $v_1,...,v_n$, which gives us an adjacency matrix of $G_1$, $A_1$. Since $f$ is a bijection, we can let the ordering of vertices of $G_2$ be $f(v_1),...,f(v_n)$, and this gives us $A_2$.
• Now in $A_1$, when $i\ne j$, ${A_1}_{ij}$ represents the number of edges incident on $v_i,v_j$, say $k$, so $e_1,...,e_k\in E_1$ are incident on $v_i,v_j$. In $A_2$, ${A_2}_{ij}$ represents the number of edges incident on $f(v_i),f(v_j)$, and since $e_1,...,e_k\in E_1$ are incident on $v_i,v_j$, $g(e_1),...,g(e_k)\in E_2$ must also be incident on $f(v_i),f(v_j)$ by the definition of isomorphism.
• Alternatively, we can argue that ${A_2}_{ij}=l$, then $e_1^{\prime },...,e_l^{\prime }\in E_2$ are incident on $f(v_i),f(v_j)$. Then, we use the fact that since $g$ is a bijection, the inverse $g^{-1}:E_2\to E_1$ exists and is a bijection. Therefore, $g^{-1}(e_1^{\prime }),...,g^{-1}(e_l^{\prime })\in E_1$ are incident on $v_i,v_j\in V_1$, which means $l=k$.
• For the diagonal entries, the proof is essentially the same, but be careful when writing the numbers because you have to double the number of loops in those entries.
• Thus, we found such orderings so that the adjacency matrices are exactly the same.
  Ordering$⟹$Isomorphism
• Suppose there exist orderings of vertices in $G_1$ and vertices of $G_2$ such that $A_1=A_2$. As $A_1$ and $A_2$ are equal, they are equal in size, so $G_1,G_2$ have the same number of vertices. Let the ordersing be $v_1,..,v_n\in V_1$ and $v_1^{\prime },...,v_n^{\prime }\in V_2$ respectively.
• Define $f:V_1\to V_2$ by $f(v_1)=v_1^{\prime },..,f(v_n)=v_n^{\prime }$. $F$ is a function because it is defined everywhere on $V_1$, and each $v_i\in V_1$ corresponds to a unique $v_i^{\prime }\in V_2$ as the vertices are all distinct. Moreover, $f$ is injective because if $i\ne j$, then $f(v_i)\ne f(v_j)$ as the vertices are distinct; $f$ is also surjective because if $v_i^{\prime }\in V_2$, then due to how we defined the function in the first place, $f(v_i)=v_i^{\prime }$. Thus, $f$ is a bijection.
• We then define $g:E_1\to E_2$. When $i\ne j$, suppose ${A_1}_{ij}=k$, then $\exists$ distinct $e_1,..,e_k\in E_1$ that are incident on $v_i,v_j\in V_1$, and because $A_1=A_2$, ${A_2}_{ij}=k$, so $\exists$ distinct $$e_1^{\prime },..,e_k^{\prime }\in E_2$ that are incident on $v_i^{\prime }=f(v_i),v_j^{\prime }=f(v_j)$. Thus, we can define $g(e_1)=e_1^{\prime },..,g(e_k)=e_k^{\prime }$. To avoid re-definitions, we restrict $i=1,...,n-1$ and $j=2,...,n$ (i.e. all entries above the main diagonal).
• When $i=j$, if ${A_1}_{ij}=l$, we know we have distinct loops $e_1,..,e_{\frac{l}{2}}\in E_1$ incident on $v_i\in V_1$, and thus we also have distinct $e_1^{\prime },...,e_{\frac{l}{2}}^{\prime }\in E_2$ incident on $v_i^{\prime }=f(v_i)\in V_2$. Thus, we define $g(e_1)=e_1^{\prime },...,g(e_{\frac{l}{2}})=e_{\frac{l}{2}}^{\prime }$.
• $g$ is a bijection naturally from our definition.
• Thus, $(f,g)$ gives an isomorphism, which establishes the proposition.

Remark

• This propistion allow us to prove or disprove two graphs are isomorphic by enumerating all their adjacency matrices and see if we find an euqal pair or not.

Isomorphism for Simple Graphs

We will be using the proposition proved above to prove the following theorem about simple graphs’ isomorphism.

Theorem

• Let $G_1$ and $G_2$ be simple graphs. The following are equivalent:
  • (a) $G_1$ and $G_2$ are isomorphic.
  • (b) There is a one-to-one, onto function f from the vertex set of $G_1$ to the vertex set of $G_2$ satisfying the following: Vertices $v$ and $w$ are adjacent in $G_1$ if and only if the vertices $f(v)$ and $f(w)$ are adjacent in $G_2$ .

Proof

(a)$⟹$(b)

• Follows the definition of graph isomorphism.

(b)$⟹$(a)

• Suppose we have such a bijection $f$. It implies that $|V_1|=|V_2|$ because a bijection between two (finite) sets exists if and only if the sets have the same cardinality (by the definition of two sets having the same cardinality). Therefore, we can choose the ordering of $V1$ to be $(v_1,...,v_n)$ and thus $(f(v_1),...,f(v_n))$ is an ordering of $V_2$.
• Let $A_1$ be the adjacency matrix of $G_1$ relative to the first ordering stated above and $A_2$ of $G_2$ to the second. Because there is no loops or parallel edges in $G_1,G_2$, each entry in any of the adjacency matrices for $G_1$ and $G_2$ is either $0$ or $1. Thus, since $f$ satisfies vertices $v$ and $w$ are adjacent in $G_1$ if and only if the vertices $f(v)$ and $f(w)$ are adjacent in $G_2$, $A_1=A_2$, so isomorphism follows the proposition.