next up previous
Next: About this document ... Up: Graphs Previous: Automorphisms of finite projective

Symmetry and connectivity of graphs

Definition 3.6   An undirected graph $ X=(V,E)$ is connected if every pair of vertices is connected by a path. It is $ k$-connected if it remains connected after removing any $ k-1$ vertices (except the complete graph $ K_n$ which is $ n-1$-connected by convention). Let $ \kappa(X)$ be the maximum $ k$ such that $ X$ is $ k$-connected. $ \kappa(X)$ is called the vertex-connectivity of $ X$.

Example 3.7   The cycle graph has $ \kappa(C_n)=2$ for all $ n\ge 3$. One should think of this as a topological invariant and all cycles have the same shape. This requires the convention that $ \kappa(C_3)=\kappa(K_3)=2$.

Definition 3.8   If $ s,t$ are vertices in a graph $ X$, then the $ s$-$ t$-connectivity $ \kappa(X;s,t)$ is the minimum number of vertices to delete to prevent there from being a path from $ s$ to $ t$. (If there's a direct edge from $ s$ to $ t$, count deleting it as deleting a vertex.) Then $ \kappa(X)=\min_{s,t}\{\kappa(X;s,t)\}$. There is a similar notion $ \kappa^+(X;s,t)$ for directed graphs and directed paths from $ s$ to $ t$. There is also the edge $ s$-$ t$-connectivity $ \rho(X;s,t)$, the minimum number of edges to delete to prevent there from being a path from $ s$ to $ t$. This has a global version $ \rho(X)=\min_{s,t}\{\rho(X;s,t)\}$ and a directed version $ \rho^+$.

Theorem 3.9 (Menger)   $ \kappa(X;s,t)$ is also the number of vertex-disjoint paths from $ s$ to $ t$. $ \rho(X;s,t)$ is also the number of edge-disjoint paths from $ s$ to $ t$. The analogous directed versions are also true.


\begin{exercise}
Assume we know the directed-edge-version of Menger's Theorem.
Deduce the other three versions.
\end{exercise}


\begin{exercise}
% latex2html id marker 877
(Fundamentals of Combinatorial Duali...
...rem of Linear Programming.
Ask about these theorems in tutorial.
\end{exercise}


\begin{exercise}
% latex2html id marker 879
$\kappa\le \rho\le \min_{v\in V}\deg(v)$.
\end{exercise}

Definition 3.10   $ X$ is a vertex-transitive graph if its automorphism group acts transitively on its vertices. Similarly, it is edge-transitive if the automorphism group acts transitively on the edges. It is vertex- or edge-primitive if the automorphism group acts primitively on the set of vertices (edges, respectively).


\begin{exercise}
% latex2html id marker 881Every vertex transitive graph is regular (all vertices have the same
degree).
\end{exercise}


\begin{exercise}
% latex2html id marker 883
Construct a graph which is vertex-transitive but not edge-transitive.
\end{exercise}


\begin{exercise}
% latex2html id marker 885
(easy) Construct a graph which is edge-transitive but not
vertex-transitive.
\end{exercise}


\begin{exx}
% latex2html id marker 887
Construct a graph which is edge-transitive, not vertex-transitive,
but is {\bf regular}.
\end{exx}

Theorem 3.11 (Mader-Watkins)   If $ X$ is an undirected, connected, vertex-transitive graph, regular of degree $ d$, then
  1. $ \kappa(X)\ge\lceil\frac{2d+1}3\rceil$;
  2. $ \rho(X)=d$;
  3. if $ X$ is edge-transitive or vertex-primitive, then $ \kappa(X)=d$.


\begin{exercise}
% latex2html id marker 889Show that $\lceil\frac{2d+1}3\rceil...
...$.
That is, construct graphs with exactly that value of $\kappa$.
\end{exercise}

Definition 3.12   A directed graph is strongly connected if for every pair of vertices $ s$ and $ t$, there is a directed path from $ s$ to $ t$. It is weakly connected if it is a connected as an undirected graph when we ignore the orientation of the edges.

Definition 3.13   A directed graph $ X$ satisfies the Euler condition if for each vertex the in-degree is equal to the out-degree.


\begin{exercise}
% latex2html id marker 891(a)\ Prove that every finite vertex...
...hat this statemnt is false for
infinite vertex-transitive graphs.
\end{exercise}


\begin{exercise}
% latex2html id marker 893
If a directed graph $X$\ satisfies t...
...n it is strongly connected if and only if it is weakly connected.
\end{exercise}

Theorem 3.14 (Hamidoune)   If $ X$ a finite, connected vertex-transitive digraph then
  1. $ \kappa^+(X)\ge\lceil\frac{d+1}2\rceil$;
  2. $ \rho^+(X)=d$;
  3. if $ X$ is edge-transitive or vertex-primitive, then $ \kappa^+(X)=d$.


\begin{exercise}
% latex2html id marker 895
Use Hamidoune's theorem to prove the...
...enport Theorem.
\\ {\em Hint.}\ Use the vertex-primitive version.
\end{exercise}




A typesetting command error (a signle omitted backslash) rendered a paragraph unintelligible in the previous handout. It was the paragraph connecting two theorems. Here we reproduce the two theorems with the corrected paragraph inbetween.


Theorem 3.15 (B-Seress)   Let $ S$ generate $ S_n$. % latex2html id marker 2179
$ \mathop{\rm diam}\nolimits (S_n,S)< m_n^{1+o(n)}$.

This result suffers from the ``element-order bottleneck.'' The two tricks used are commutators and raising elements to powers. A recent result gives a polynomial upper bound under the condition that one of the generators fixes 70% of the permutation domain. So now all is left to prove is that we can reach such a permutation in a polynomially bounded number of steps. Somebody in this audience may be able to do this with a fresh idea.

Theorem 3.16 (B-Beals-Seress)   Let $ S$ generate $ S_n$. If % latex2html id marker 2186
$ \exists s\in S$ with the % latex2html id marker 2188
$ \deg s<0.3n$ then % latex2html id marker 2190
$ \mathop{\rm diam}\nolimits (S_n,S)<n^C$ ($ C=12$?). (Recall that the degree of a permutation is size of its support, the set of elements that it actually moves.)


next up previous
Next: About this document ... Up: Graphs Previous: Automorphisms of finite projective
Varsha Dani 2003-08-04