Discrete Mathematics – MATH 1006

Lecture 11

Lecture 11 Discrete Mathematics – MATH 1006 1 / 33

Lecture outline
Last week
Rational numbers
Recursion and induction
Catalan numbers again

This week
Graph theory:
Graphs
Edge-endpoint functions
Degree
Walks and paths

Lecture 11 Discrete Mathematics – MATH 1006 2 / 33

Graph Theory

In school, you studied graphs like:

y y= 1 2
12 x

x
These are not the kind of graphs we are studying here.

Lecture 11 Discrete Mathematics – MATH 1006 3 / 33

 Graph Theory DIAGRAM MONOIDS AND GRAHAM–HOUGHTON GRAPHS 29

It follows that the graph Λn has vertex set {τij : 1 ≤ i < j ≤ n} with edges τij → τkl if
Number Theoryand only if {i, j} ∩ {k, l} ̸= ∅. The graphs Λ4 and Λ5 are pictured in Figure 11, where
In discrete mathematics, graphs look like:
we have simplified matters by labelling the vertices ij instead of τij , omitting loops, and
displaying pairs of directed edges ij ! kl as single undirected edges ij − kl.
The shape is a “tesseract”, a 4D cube:
PB n

Bn Mn In

Idempotent generators — Pn \ Sn
Jn Sn On

32 DIAGRAM MONOIDS AND GRAHAM–HOUGHTON GRAPHS

I Minimal (idempotent) generating sets (E and Gray, 2014).
. {1}
Lecture 9 12
Discrete Mathematics – 200025 1
Figure 11. Simplified illustrations of the graphs Λ4 = Γ(B4 \ S4 ) and Figure 2: Important submonoids of PB n (left) and repr
Λ5 = Γ(B5 \ S5 ); see text for further details. See text for12further explanation.
15
Recall that the Johnson graph J(n, k) is the graph with vertex set {A ⊆ [n] : |A| = k},
and with edges A − B if and23only if |A ∩ B| = k − 1. In particular
13 J(n, 2) has 5vertex denotes
25 the identity2 Brauer diagram. The relevant par
set {A ⊆ [n] : |A| = 2} and edges A − B if and only if A and B overlap in precisely one Figure 2.13
14
element. More background on Johnson graphs in the context of algebraic graph theory
may be found in [?, Chapter 1.6]. Note that the underlying undirected loop-free graph Green’s relations on PB n (and all the submonoids mentio
of Λn = Γ(Bn \ Sn ) is isomorphic to J(n, 2). So, in fact, Figure 11 pictures the Johnson 45 35 a number
of 24 23
of parameters we now describe. With this
24 14
graphs J(4, 2) and J(5, 2). transversal block if it has nonempty intersection with both
Factorizations of Johnson graphs have been considered, for instance, in [?]. Recall that The rank of ↵, denoted rank(↵), is equal to the number
a 1-factor of a graph is a collection of edges that spans the graph, while a 2-factor is a [x]↵34denote the
4 3 block of ↵ containing x. We define the d
collection of cycles that spans all vertices of the graph. We define a (0, 1, 2)-factor of a
.
graph as a decomposition of the graph into vertices, edges, and cycles, such that each vertex
34
dom(↵) = x 2 [n] : [x]↵ \ [n]0 6= ; and
is contained in precisely one of these pieces. An oriented (0, 1, 2)-factor is a (0, 1, 2)-factor
There are
such that all the
Figure 13.cycles
Proposition 8.7. There
many
The are assigned
graph
F =is {τ
variations.
anwhere
Λ4 (F ),
a one-one I P5 \ the
, τ , σ correspondence
But
, σ , σS5, minimal
, σ , σ , σ between σ= ht
the
,}.t13 ,basic
orientation (clockwise or anticlockwise).
tidea
t15 , t24 ,We
, σ12idempotent
is:the kernel and cokernel of ↵ to be the eq
also define
34 , t45 , t4 ,
14 23 214 231 234 241 314 321 324 341
generating sets of Bn \ Sn and the oriented (0, 1, 2)-factors of the Johnson graph J(n, 2). ker(↵) = (x, y) 2 [n] ⇥ [n] : [x]↵ = [y]↵ and
vertices (i.e., nodes) joined by edgeseTo41(i.e.,
, f14 , e23lines).
, f23 , e35 , f35 , e52 , f52 i.
Proof. This is an immediate consequence of Theorem 5.7, the above-mentioned connection
illustrate these ideas, let
between Λn = Γ(Bn \Sn ) and J(n, 2), and the associated correspondence between balanced
subgraphs of Λn and oriented (0, 1, 2)-factors of James
J(n, 2).
East 31" Motzkin
"= monoids
Remark 8.8. We do not know of any formula
12 or recurrence relation for the number of
Lecture 11 Discrete Mathematics
Then–rank(")
MATH 1006
= 3, dom(") = {4, 7,48},
/ 33
codom(") = {5, 6, 8
Graph Theory

Graphs are used in many contexts:

Social media Transport networks Computer networks

Chemistry Physics Neural networks

Lecture 11 Discrete Mathematics – MATH 1006 5 / 33
Graph Theory

Even the movies:

For an explanation, see:

https://www.youtube.com/watch?v=iW_LkYiuTKE
Lecture 11 Discrete Mathematics – MATH 1006 6 / 33
Graph Theory

Definition (graph)

A graph consists of two sets:

a set V of vertices, and
a set E of edges.

Every edge has either one or two vertices specified as its

endpoints.

Lecture 11 Discrete Mathematics – MATH 1006 7 / 33

Graph Theory

The correspondence between edges and endpoints is called the

edge-endpoint function.
e.g., consider a graph G with vertex set and edge set:
V = {1, 2, 3, 4, 5} and E = {a, b, c, d}

and edge-endpoint function:

edge endpoint(s)
a 2, 3
b 1
c 2, 3
d 1, 5

The endpoints of edge a are the vertices 2 and 3.

Edge b only has one endpoint (vertex 1), etc.
Lecture 11 Discrete Mathematics – MATH 1006 8 / 33
Graph Theory

A graph may be drawn in many ways.

e.g., let G be the graph defined on the previous page:

Graph G edge endpoint(s)

a 2, 3
V = {1, 2, 3, 4, 5}
b 1
c 2, 3
E = {a, b, c, d}
d 1, 5

Then G may be drawn as:

d

a
b 1 2 3 4 5
c

Lecture 11 Discrete Mathematics – MATH 1006 9 / 33

Graph Theory

Here are some other ways to draw the same graph:

Graph G edge endpoint(s)

a 2, 3
V = {1, 2, 3, 4, 5}
b 1
c 2, 3
E = {a, b, c, d}
d 1, 5

b b

d
1 1 c 2
5 2
c a
d
a
4 3 3 5 4

Lecture 11 Discrete Mathematics – MATH 1006 10 / 33

Graph Theory

Note that a picture of a graph is not the graph.

It is just a visual representation of the graph.

A graph has many such visual representations.

But a graph is its vertex set, edge set, and edge-endpoint

function.

i.e., a graph is an abstract system of vertices, edges, and

relationships.

Pictures help us understand graphs.

Lecture 11 Discrete Mathematics – MATH 1006 11 / 33

Graph Theory
Definition (terms associated to a graph)

Consider a graph G .

An edge of G with only one endpoint is called a loop.

Two edges with the same endpoints are called parallel.

A vertex that is not the endpoint of any edge is called

isolated.

A graph with no vertices is called empty.

A graph with at least one vertex is called non-empty.

Lecture 11 Discrete Mathematics – MATH 1006 12 / 33

Graph Theory

e.g., let G be the graph defined earlier:

Graph G edge endpoint(s)

a 2, 3
V = {1, 2, 3, 4, 5}
b 1
c 2, 3
E = {a, b, c, d}
d 1, 5
d

a
b 1 2 3 4 5
c

Vertex 4 is isolated. Edges a and c are parallel.

Edge b is a loop.

Lecture 11 Discrete Mathematics – MATH 1006 13 / 33

Graph Theory

From a picture of a graph, we can give its abstract description.

e.g., suppose a graph G can be pictured as:

1
u v

3 2

5 x 4 w

Then its abstract description is:

Graph G edge endpoint(s)
1 u, v
V = {u, v , w , x} 2 v, w
3 v, w
E = {1, 2, 3, 4, 5} 4 x
5 x
Lecture 11 Discrete Mathematics – MATH 1006 14 / 33
Graph Theory

Definition (simple graph)

A simple graph is a graph with no loops and no parallel edges.

1 γ
1 u v 1 2
d

5 2 3 2 α
c δ
β
a
4 3 x w 4 3

not simple not simple simple

In a simple graph, an edge is completely determined by its

endpoints.
So edges are often not labelled in simple graphs.
Sometimes an edge is labelled by its endpoints. e.g., δ = {2, 3}.
Lecture 11 Discrete Mathematics – MATH 1006 15 / 33
Graph Theory

Exercises
1 Draw all simple graphs with vertex set {1, 2, 3, 4} and one
edge.
2 How many simple graphs have vertex set {1, 2, 3, 4} and two
edges?
3 How many simple graphs have vertex set {1, 2, 3, 4} and three
edges?
4 How many simple graphs have vertex set {1, 2, 3, 4}?
5 How many simple graphs have vertex set {1, 2, . . . , n}?

Lecture 11 Discrete Mathematics – MATH 1006 16 / 33

Graph Theory

Definition (complete graph)

The complete graph, KV , is the simple graph with:

vertex set V and all possible edges.

i.e., KV has the edge {u, v } for all u, v ∈ V with u 6= v .

When V = {1, 2, . . . , n}, we write Kn for KV .

Note: Kn has n2 = n(n−1) 2 edges. Why?

1
1 1 2
5 2

1 1 2 3 2 4 3 4 3

K1 K2 K3 K4 K5

Lecture 11 Discrete Mathematics – MATH 1006 17 / 33

Graph Theory

1 1 2
1 2
7 2
8 3

6 3
6 3
7 4

5 4 5 4 6 5

K6 K7 K8

1 1 2 1
11 2
9 2
10 3
10 3
8 3
9 4
9 4

7 4 8 5 8 5
6 5 7 6 7 6

K9 K10 K11

Lecture 11 Discrete Mathematics – MATH 1006 18 / 33

Graph Theory

K12
Lecture 11 Discrete Mathematics – MATH 1006 19 / 33
Graph Theory

Definition (degree)

Let v be a vertex in a graph G (not necessarily simple).

The degree of v is the number of edges coming out of v .

Loops are counted twice.

We write deg(v ) for the degree of v .

Graph G vertex degree

u v u 3
v 3
w 2
x w x 0

Lecture 11 Discrete Mathematics – MATH 1006 20 / 33

Graph Theory

Definition (total degree)

Let G be a graph (not necessarily simple) with vertex set V .

The total degree of G is the sum of all degrees.

X
i.e., deg(v ).
v ∈V

Graph G vertex degree

u v u 3
v 3
w 2
x w x 0
Total degree 8

Lecture 11 Discrete Mathematics – MATH 1006 21 / 33

Graph Theory
Useful Fact
The total degree of a graph is equal to twice the number of edges:
X
deg(v ) = 2 × |E |.
v ∈V

Graph G vertex degree

u v u 3
v 3
w 2
x w x 0
Total degree 8
Number of edges 4

Proof of Useful Fact: When adding up the degrees, each edge

contributes 2 to the sum. 
Lecture 11 Discrete Mathematics – MATH 1006 22 / 33
Graph Theory

Exercises
1 How many edges are there in a graph with 10 vertices, each of
degree 6?
2 Draw a graph with 4 vertices and the following degrees, or
explain why no such graph exists:
1, 1, 1, 4 1, 1, 1, 3 1, 1, 1, 1 1, 2, 3, 4

3 Draw a simple graph with 4 vertices and the following

degrees, or explain why no such simple graph exists:
1, 2, 3, 4 1, 1, 3, 3 1, 1, 1, 1 2, 2, 3, 3

4 In a group of 7 people, is it possible for each to be friends

with exactly 3 others?

Lecture 11 Discrete Mathematics – MATH 1006 23 / 33

Graph Theory — the Königsberg Bridge Problem

Graph theory was invented in 1736 by Leonhard Euler.

He invented it to solve the following problem.
The Königsberg Bridge Problem

There are seven bridges in Königsberg (see below).

Is it possible to go for a walk, starting and finishing at Euler’s

house, and using each bridge exactly once?

Lecture 11 Discrete Mathematics – MATH 1006 24 / 33

Graph Theory — the Königsberg Bridge Problem

Here is a simplified picture of the Königsberg map:

C

B D

To solve the problem, we C

define a graph.

The vertices are A, B, C , D.

A
Bridges give the edges. B D

Note: not to scale! All that matters are the connections.

Lecture 11 Discrete Mathematics – MATH 1006 25 / 33
Graph Theory — the Königsberg Bridge Problem

A
B D

The question now becomes:

Can you trace a path in the above graph,

starting and finishing at the same place,
using each edge exactly once, and
never lifting your pen?
Try it!
To solve the problem, we need to develop some more theory...
Lecture 11 Discrete Mathematics – MATH 1006 26 / 33
Graph Theory — walks, paths, circuits

Definition (walk)

A walk in a graph is an alternating sequence of vertices and

edges:
e e e e e
v0 −−1→ v1 −−2→ v2 −−3→ v3 −−4→ · · · · · · · · · −−k→ vk .

The vertices on either side of an edge must be its endpoints.

If the first vertex of a walk is u, and the last is v , then we say

it is a walk from u to v .

Lecture 11 Discrete Mathematics – MATH 1006 27 / 33

Graph Theory — walks, paths, circuits

Consider the following graph:

a

b c
1 2 3

g d
e
h 5 4
f
b c d e h h f
1 −−→ 2 −−→ 3 −−→ 4 −−→ 5 −−→ 5 −−→ 5 −−→ 4
is a walk from 1 to 4.
g h f d c b a
2 −−→ 5 −−→ 5 −−→ 4 −−→ 3 −−→ 2 −−→ 1 −−→ 1
is a walk from 2 to 1.
Lecture 11 Discrete Mathematics – MATH 1006 28 / 33
Graph Theory — walks, paths, circuits

Special types of walk

A walk with no edges is called trivial.

A walk with at least one edge is called non-trivial.

A walk with no repeated edges is called a path.
Note: a path might repeat vertices.

A path from a vertex to itself is called a circuit.

A path that includes every edge is called an Euler path.
Note: an Euler path uses each edge exactly once.

An Euler path that is also a circuit is called an Euler circuit.

Note: an Euler circuit uses each edge exactly once, and returns
to where it started.
Lecture 11 Discrete Mathematics – MATH 1006 29 / 33
Graph Theory — walks, paths, circuits

So the Königsberg Bridge Problem becomes:

Question
Does the following graph have an Euler circuit?

B D

Before we answer this, let’s consider another example...

Lecture 11 Discrete Mathematics – MATH 1006 30 / 33
Graph Theory — walks, paths, circuits

Consider the following graph:

b
2 3
a h c

1 g j 4

f i d
6 e 5
g i j
6 −−→ 2 −→ 5 −→ 3

is a path (not a circuit, not Euler).

g i e h j e
6 −−→ 2 −→ 5 −−→ 6 −−→ 3 −→ 5 −−→ 6

is not a path (so not a circuit, Euler, etc).

Lecture 11 Discrete Mathematics – MATH 1006 31 / 33
Graph Theory — walks, paths, circuits

Consider the following graph:

b
2 3
a h c

1 g j 4

f i d
6 e 5
g i e
6 −−→ 2 −→ 5 −−→ 6 is a circuit (not Euler).

g i e h j d c b a f
6 −→ 2 −→ 5 −→ 6 −→ 3 −→ 5 −→ 4 −→ 3 −→ 2 −→ 1 −→ 6

is an Euler circuit!

Lecture 11 Discrete Mathematics – MATH 1006 32 / 33

Graph Theory — walks, paths, circuits

Question (to be answered next time)

Does Euler’s graph have an Euler circuit?

B D

Lecture 11 Discrete Mathematics – MATH 1006 33 / 33