Diskrete Mathematik HS 13 Definition 2.5.
(i) The negation (logical NOT) of a proposition A, denoted
Zusammenfassung
2. Mathematical Reasoning, Proofs, and a
First Approach to Logic
2.1 What is a Proof?
Definition 2.1. (Informal.) A proof of a statement S is a se-
quence of simple, easily verifiable, consecutive steps. The
proof starts from a set of axioms (things postulated to be true)
and known (previously proved) facts. Each step corresponds
to the application of a derivation rule to a few already proven
statements, resulting in a newly proved statement, until the
final step proves S.
2.2 Propositions and Logical Formulas
Definition 2.2. A proposition [Aussage] is a (mathematical)
statement that is either true or false.
Definition 2.3. A true proposition is often called a theorem, a
lemma, or a corollary.1
Definition 2.4. The logical values (constants) “true” and
“false” are usually denoted as 1 and 0, respectively.
1 In mathematical texts one sometimes calls a true statement a proposition
(instead of a theorem). In practice this will not be confusing.
1 2
Definition 2.10. A formula F (in propositional logic) is called
satisfiable [erfüllbar] if it is true for at least one truth assign-
ment of the involved propositional symbols, and it is called
unsatisfiable otherwise.
Lemma 2.1. F is a tautology iff ¬F is unsatisfiable.
Definition 2.11. One writes F =⇒ G (or F ⇒ G) to say that
F implies G, i.e., that F → G is a tautology.
Theorem 2.2. Implication is transitive: If F =⇒ G and G =⇒ H,
then F =⇒ H.
2.3 Quantifiers and Predicate Logic
Let us consider a set U as the universe in which we want to
reason.
Definition 2.12. A k-ary predicate [Prädikat] P on U is a func-
tion U k → {0, 1}.
Definition 2.13. For a universe U and predicate P (x) we de-
fine the following logical statements:3
∀x P (x) is the statement that P (x) is true for all x ∈ U .
∃x P (x) is the statement that P (x) is true for some x ∈
U , i.e., there exists an x ∈ U for which P (x) is true.
3 Inthe literature one also finds the notations ∀x : P (x) and ∀x. P (x)
instead of ∀x P (x), and similarly for ∃.
3 4
as ¬A, is true iff A is false.
(ii) The conjunction (logical AND) of two propositions A
and B, denoted A ∧ B, is true iff both A and B are true.
(iii) The disjunction (logical OR) of two propositions A and
B, denoted A ∨ B, is true iff A or B (or both) are true.
Definition 2.6. A correctly formed expression involving
propositional symbols (like A, B, C, . . .) and logical operators
is called a formula (of propositional logic).
We introduce a new, derived logical operator: implication,
denoted as A → B and defined by2 A → B :⇐⇒ ¬A ∨ B.
Two-sided implication, denoted A ↔ B, is defined as follows:
A ↔ B :⇐⇒ (A → B) ∧ (B → A).
There are also priority rules for logical operators which allow
us to simplify the notation, in the same sense as in algebra one
can write ab + c (rather than (a · b) + c). Namely, ∧ and ∨ bind
stronger than → and ↔. Also, ¬ binds stronger than ∧ and ∨.
Definition 2.7. The symbol > denotes the function that is the
constant 1 (true), and ⊥ denotes the function that is constant
0 (false).
Definition 2.8. Two formulas F and G (in propositional logic)
are called equivalent, denoted as F ⇐⇒ G (or also as F ≡ G),
if they correspond to the same function (table).
Definition 2.9. A formula F (in propositional logic) is called
a tautology [Tautologie] if it is true for all truth assignments
of the involved propositional symbols.
2 The symbol :⇐⇒ simply means that the left side (here A → B) is defined
to mean the right side (here ¬A ∨ B).
2.4 Some Proof Patterns and Techniques
Theorem 2.3. If F and F → G are tautologies, then G is also a
tautology.
Definition 2.14. A direct proof of an implication F → G
works by assuming F and then deriving G (from F ), where
the derivation can possibly involve several proof steps.
Definition 2.15. An indirect proof of an implication F → G
works by assuming ¬G and deriving ¬F , i.e., by proving
¬G → ¬F .
Theorem 2.4. If ¬F → G and ¬G are tautologies, then F is
also a tautology.
Definition 2.16. An existence proof is the proof of a statement
of the form ∃x P (x).
Definition 2.17. An inexistence proof is a proof of a statement
of the form ¬∃x P (x).
Definition 2.18. A proof by counterexample is a proof of a
statement of the form ¬∀x P (x) for some fixed predicate P ,
using ¬∀x P (x) ⇐⇒ ∃x ¬P (x). An a for which ¬P (a) is
true is called a counterexample.
A proof by induction consists of two steps:
Proof by induction:
1. Basis step. Prove P (0).
2. Induction step. Prove ∀n (P (n) → P (n + 1)).
Theorem 2.5. For every predicate P on N we have P (0) ∧
∀n (P (n) → P (n + 1)) =⇒ ∀n P (n).
3. Sets, Relations, and Functions
3.1 Sets and Operations on Sets
Definition 3.1. A = B :⇐⇒ ∀x (x ∈ A ↔ x ∈ B).
Definition 3.2. The number of elements of a finite set A is
called its cardinality and is denoted |A|.
A set can also be described by a defining property of its ele-
ments. If A is a set and P is a predicate defined on A, then
{x ∈ A| P (x)} (or alternatively {x ∈ A : P (x)}) denotes the
set of elements of A having property P .
For the operation of forming an ordered pair of two objects a
and b, denoted (a, b), we have (a, b) = (c, d) :⇔ a = c ∧ b = d.
Definition 3.3. The set A is a subset of the set B, denoted
A ⊆ B, if every element of A is also an element of B, i.e.,
A ⊆ B :⇐⇒ ∀x (x ∈ A → x ∈ B).
It follows directly from the definition of set equality that
A = B ⇐⇒ (A ⊆ B) ∧ (B ⊆ A).
Definition 3.4. The empty set, denoted ∅ or {}, is the set with
no elements, i.e., ∀x (x 6∈ ∅).1
Lemma 3.1. The empty set is a subset of every set, i.e., ∀A (∅ ⊆ A)
Lemma 3.2. The empty set is unique.
1 One can consider ∀x (x ∈
/ ∅) as the definition of ∅.
5 6
Distributive: A ∩ (B ∪ C) = (A ∩ B) ∪ (A ∩ C);
A ∪ (B ∩ C) = (A ∪ B) ∩ (A ∪ C);
Complementarity: A ∩ A = ∅ and A ∪ A = U ;
Consistency: A ⊆ B ⇐⇒ A ∩ B = A ⇐⇒ A ∪ B = B.
Definition 3.9. The Cartesian product A×B of two sets A and
B is the set of all ordered pairs with the first component from
A and the second component from B: A × B = {(a, b)| a ∈
A ∧ b ∈ B}.
More generally, the Cartesian product of k sets A1 , . . . , Ak is
the set of all lists of length k with the ith component from Ai :
× k
i=1 Ai = {(a1 , . . . , ak )| ai ∈ Ai for 1 ≤ i ≤ k}
3.2 Relations
Definition 3.10. A (binary) relation ρ from a set A to a set B
is a subset of A × B. If A = B, then ρ is called a relation on A.
Instead of (a, b) ∈ ρ one usually writes a ρ b,
Definition 3.11. For any set A, the identity relation on A, de-
noted id, is the relation id = {(a, a)| a ∈ A}.
Definition 3.12. The inverse of a relation ρ from A to B is the
relation ρb from B to A such that a ρ b ⇐⇒ b ρb a.
Definition 3.13. Let ρ be a relation from A to B and let σ be
a relation from B to C. Then the composition of ρ and σ, de-
noted ρσ (or also ρ ◦ σ), is the relation from A to C where
a ρσ c :⇐⇒ ∃b ∈ B : (a ρ b ∧ b σ c). The n-fold composi-
tion of a relation ρ on a set A is denoted ρn .
7 8
Definition 3.5. The power set of a set A, denoted P(A) , is the
set of all subsets of A: P(A) := {S| S ⊆ A}.
Definition 3.6. The union of two sets A and B is defined as2
A ∪ B := {x| x ∈ A ∨ x ∈ B}, and their intersection is defined
as A ∩ B := {x| x ∈ A ∧ x ∈ B}.
Let A be a set of sets. Then we define the union of all sets in
A as the set
S of all x that are an element of at least one of the
sets in A: A := {x| ∃A ∈ A : x ∈ A}. Similarly, we define
the intersection of all sets T
in A as the set of all x that are an
element of every set in A: A := {x| ∀A ∈ A : x ∈ A}.
Definition 3.7. For a given universe of discourse, U , the com-
plement of a set A, denoted A is A := {x ∈ U | x ∈ / A} or
simply A = {x| x ∈/ A}.3
Definition 3.8. The difference of sets B and A, denoted B − A
(or sometimes B \ A) is the complement of A, relative to B:
B − A := {x ∈ B| x ∈ / A}.
Theorem 3.3. For any sets A, B, and C, and a universe U , the
following laws hold:
Idempotent: A ∩ A = A and A ∪ A = A;
Commutative: A ∩ B = B ∩ A and A ∪ B = B ∪ A;
Associative: A ∩ (B ∩ C) = (A ∩ B) ∩ C and
A ∪ (B ∪ C) = (A ∪ B) ∪ C;
Absorption: A ∩ (A ∪ B) = A and A ∪ (A ∩ B) = A;
2 This definition could also be written as ∀x (x ∈ A ∪ B ↔ x ∈ A ∨ x ∈
B).
3 The complement of A is sometimes also denoted as Ac .
Lemma 3.4. The composition of relations is associative, i.e., we
have ρ(σφ) = (ρσ)φ.
Lemma 3.5. Let ρ be a relation from A to B and let σ be a relation
from B to C. Then the inverse ρc
σ of ρσ is the relation σ
bρb.
Definition 3.14. A relation ρ on a set A is called reflexive if
a ρ a for every a ∈ A, i.e., if id ⊆ ρ.
Definition 3.15. A relation ρ on a set A is called irreflexive if
a6 ρ a for all a ∈ A.
Definition 3.16. A relation ρ on a set A is called symmetric if
ρ = ρb, i.e., if a ρ b ⇐⇒ b ρ a.
Definition 3.17. A relation ρ on a set A is called antisymmet-
ric if ρ ∩ ρb ⊆ id, i.e., a ρ b ∧ b ρ a =⇒ a = b.
Definition 3.18. A relation ρ on a set A is called transitive if
a ρ b ∧ b ρ c =⇒ a ρ c.
Lemma 3.6. A relation ρ is transitive iff ρ2 ⊆ ρ.
Definition 3.19. The transitive
S∞ closure of a relation ρ on a set
A, denoted ρ∗ , is ρ∗ = n=1 ρn .
3.3 Equivalence Relations
Definition 3.20. An equivalence relation is a relation that is
reflexive, symmetric, and transitive.
Definition 3.21. For an equivalence relation θ on a set A and
for a ∈ A, the set of elements of A that are equivalent to a is
called the equivalence class of a and is denoted as [a]θ : [a]θ :=
{b ∈ A| b θ a}.
Lemma 3.7. The intersection of two equivalence relations is an
equivalence relation.
Definition 3.22. A partition of a set A is a set {Si ⊆ A}i∈I
of mutually disjoint
S subsets of A that cover A, i.e., Si ∩ Sj =
∅ for i 6= j and i∈I Si = A.
Definition 3.23. The set of equivalence classes of an equiva-
lence relation θ, denoted by A/θ := {[a]θ | a ∈ A}, is called the
quotient set of A by θ, or simply A modulo θ, or A mod θ.
Theorem 3.8. The set A/θ of equivalence classes of an equivalence
relation θ on A is a partition of A.
3.4 Partial Order Relations
Definition 3.24. A partial order on a set A is a relation that
is reflexive, antisymmetric, and transitive. A set A together
with a partial order  on A is called a partially ordered set
(or simply poset) and is denoted as (A; ).
For a partial order relation  we can define the relation a ≺ b
similar to how the relation < is obtained from ≤: a ≺ b :⇐⇒
a  b ∧ a 6= b.
Definition 3.25. For a poset (A; ), two elements a and b are
called comparable [vergleichbar] if a  b or b  a; otherwise
they are called incomparable.
Definition 3.26. If any two elements of a poset (A; ) are
comparable, then A is called totally ordered (or linearly or-
dered) by .
Definition 3.27. A poset (A; ) is well-ordered [wohlgeord-
net] if it is totally ordered and if every non-empty subset of A
has a least element.
9 10
Definition 3.33. Let (A; ) be a poset. If a and b (i.e., the set
{a, b} ⊆ A) have a greatest lower bound, then it is called the
meet of a and b, often denoted a ∧ b. If a and b have a least up-
per bound, then it is called the join of a and b, often denoted
a ∨ b.
Definition 3.34. A poset (A; ) in which every pair of ele-
ments has a meet and a join is called a lattice [Verband].
3.5 Functions
Definition 3.35. A function f : A → B from a domain [Def-
initionsbereich] A to a codomain [Bildbereich, Wertebereich]
B is a relation on A × B, i.e., a subset f ⊆ A × B, with the
special properties (using the relation notation a f b):
1. ∀a ∈ A ∃b ∈ B a f b (f is totally defined),
2. a f b ∧ a f b0 ⇒ b = b0 (f is well-defined).
The set of all functions A → B is denoted as B A .
Definition 3.36. A partial function is a relation on A×B such
that condition 2. above holds.
Definition 3.37. Two (partial) functions with common do-
main A and codomain B are equal if they are equal as rela-
tions (i.e., as sets).
f = g is equivalent to saying that the function values of f
and g agree for all arguments (including, in case of partial
functions, whether or not it is defined).
Definition 3.38. For a function f : A → B and a subset S
of A, the image [Bild] of S under f , denoted f (S), is the set
f (S) := {f (a) | a ∈ S}.
11
Definition 3.28. In a poset (A; ) an element b is said to cover
[überdecken] an element a if a ≺ b and there exists no c with
a ≺ c and c ≺ b (i.e., between a and b).
Definition 3.29. The Hasse diagram of a (finite) poset (A; )
is the directed graph whose vertices are labeled with the ele-
ments of A and where there is an edge from a to b iff b covers
a.
The Hasse diagram is a graph with directed edges. It is usu-
ally drawn such that whenever a ≺ b, then b is placed higher
than a. This means that all arrows are directed upwards and
therefore can be omitted.
Theorem 3.9. For given posets (A; ) and (B; v), the relation ≤
defined on A×B by (a1 , b1 ) ≤ (a2 , b2 ) :⇐⇒ a1  a2 ∧ b1 v b2
is a partial order relation.
Theorem 3.10. For given posets (A; ) and (B; v), the relation
≤lex defined on A × B by (a1 , b1 ) ≤lex (a2 , b2 ) :⇐⇒ a1 ≺
a2 ∨ (a1 = a2 ∧ b1 v b2 ) is a partial order relation.
Definition 3.32. Let (A; ) be a poset, and let S ⊆ A be some
subset of A. Then
1. a ∈ S is a minimal (maximal) element of S if there exists
no b ∈ S with b ≺ a (b  a).
2. a ∈ S is the least (greatest) element of S if a  b (a  b)
for all b ∈ S.
3. a ∈ A is a lower (upper) bound [untere (obere) Schranke]
of S if a  b (a  b) for all b ∈ S.
4. a ∈ A is the greatest lower bound (least upper bound) of
S if a is the greatest (least) element of the set of all lower
(upper) bounds of S.
Definition 3.39. The subset f (A) of B is called the image (or
range) of f and is also denoted Im(f ).
Definition 3.40. For a subset T of B, the inverse image (or
preimage) [Urbild] of T , denoted f −1 (T ), is the set of values
in A that map into T : f −1 (T ) := {a ∈ A| f (a) ∈ T }.
Definition 3.41. A function f : A → B is called
1. injective if a 6= b ⇒ f (a) 6= f (b), i.e., no two distinct
values are mapped to the same function value (there are
no “collisions”).
2. surjective (or onto) if for every b ∈ B, b = f (a) for some
a ∈ A, i.e., if f (A) = B (every value in the codomain is
taken on for some argument).
3. bijective if it is both injective and surjective.
Definition 3.42. The composition of a function f : A → B
and a function g : B → C, denoted by g ◦ f or simply gf , is
defined by g ◦ f (a) = g(f (a)).
Lemma 3.11. Function composition is associative, i.e., (h◦g)◦f =
h ◦ (g ◦ f ).
12
4. Combinatorics and Counting Ordered Unordered
With n+k−1
nk
4.1 Basic Counting Principles repetition   k
Without n! n n!
:=
The cardinality of the union of n disjoint sets repetition (n−k)! k k! (n − k)!
A1 , . . . , An is equal to the sum of the cardinal-
ities. This is known as the addition principle: Table 4.1: Drawing k from n elements.
∀i,
Pn j, 1 ≤ i < j ≤ n : Ai ∩ Aj = ∅ =⇒ |A1 ∪ · · · ∪ An | =
i=1 |A i |.
Consider the problem of counting a subset S of A × B . We
The multiplication principle for finite sets states the ob- can count S in two different ways, either by determining for
vious fact that |AQ× B| = |A| · |B| and, more generally, each a ∈ A the number ma of b ∈ B such that (a, b) ∈ S, or by
n
|A1 × · · · × An | = i=1 |Ai |. determining for each bP∈ B the number
P nb of a ∈ A such that
(a, b) ∈ S. Then |S| = a∈A ma = b∈B nb . This is known as
the double-counting principle.
Bijection principle: If there is a bijection (or one-to-one cor-
respondence) between the finite sets A and B, then |A| = |B|. Theorem 4.3. If a set of n objects is partitioned into k < n sets,
then at least one of these sets contains at least d nk e objects.
Theorem 4.1. For any finite sets A1 , . . . , An ,
n
X X
4.2 Binomial Coefficients
|A1 ∪ · · · ∪ An | = |Ai | − |Ai1 ∩ Ai2 |


n n
i=1 1≤i1 <i2 ≤n Lemma 4.4. k = n−k .
X
+ |Ai1 ∩ Ai2 ∩ Ai3 |



n n−1 n−1
1≤i1 <i2 <i3 ≤n Lemma 4.5. For n > 0, k = k−1 + k .
n−1
− · · · + (−1) |A1 ∩ · · · ∩ An |.
Theorem 4.6. For any real (or complex)
Pn numbers

x and y and for
n n−k k
n
X X every integer n ≥ 0, (x + y)n = k=0 k x y .
Theorem 4.2. |A1 ∪ · · · ∪ An | ≥ |Ai | − |Ai1 ∩ Ai2 |.
i=1 1≤i1 <i2 ≤n
Corollary 4.8. Let k, m, n ≥ 0 be integers with m + n > 0. Then,
m+n

Pk m

n
2n

Pn

n 2
We recall that k! = 1 · 2 · 3 · · · (k − 1) · k (k! is called “k k = i=0 i k−i . In particular, n = k=0 k .
factorial”).
13 14

4.4 Countable and Uncountable Sets Corollary 4.14. The direct product A × B of two countable sets A
and B is countable, i.e., A  N ∧ B  N =⇒ A × B  N.
Definition 4.2.
(i) Two sets A and B have the same cardinality, denoted Corollary 4.15. The rational numbers Q are countable.
A ∼ B, if there exists a bijection A → B.
Theorem 4.16. Let A and A1 , A2 , . . . be countable sets.
(ii) The cardinality of B is at least the cardinality of A, de-
(i) For any n ∈ N, the set An of n-tuples over A is countable.
noted A  B, if A ∼ C for some subset C ⊆ B.
(iii) B dominates A, denoted A ≺ B, if A  B and A 6∼ B. (ii) The union A1 ∪ A2 ∪ · · · of a countable list of countable sets
is countable.
(iv) A set A is called countable [abzählbar] if A  N, and
uncountable [überabzählbar] otherwise. (iii) The set A∗ of finite sequences over A is countable.

Definition 4.3. Let {0, 1}∞ denote the set of semi-infinite bi-
Lemma 4.10. nary sequences.
(i) The relation ∼ is an equivalence relation.1 Theorem 4.17. The set {0, 1}∞ is uncountable.
(ii) The relation  is transitive: A  B ∧ B  C =⇒ A  C.
Lemma 4.19. If A is uncountable and A  B, then B is uncount-
(iii) A ⊆ B =⇒ A  B. able, i.e., A 6 N ∧ A  B =⇒ B 6 N. In particular, if a subset
(iv) A subset of a countable set is also countable: A ⊆ B ∧ B  of a set B is uncountable, then so is B.
N =⇒ A  N.
Lemma 4.20. If A is uncountable and B is countable, then A − B
(v) A  B ∧ B  A =⇒ A ∼ B. is uncountable.
(vi) For two sets A and B, exactly one of A ≺ B, A ∼ B, and
B ≺ A holds. Theorem 4.21. The set R of real numbers is uncountable.

Theorem 4.11. A set A is countable iff it is finite or if A ∼ N. Lemma 4.22. The interval [0, 1) of the real numbers is uncount-
able.
Theorem 4.12. The set {0, 1}∗ := {, 0, 1, 00, 01, 10, 11, 000, . . .}
of finite binary sequences is countable.2
Theorem 4.13. The set N × N (= N2 ) of ordered pairs of natural
numbers is countable.
1 Here ∼ and  should be understood as relations on a set of sets.
2 Here  denotes the empty string.
15 16
5. Graph Theory
5.2 Basic Concepts
Definition 5.1. A (simple) graph G = (V, 
E) consists of a finite
set V of vertices [Knoten] and a set E ⊆ {u, v} ⊆ V | u 6= v
of edges [Kanten].
An edge {u, v} ∈ E is said to connect the vertices u and v.
Vertices connected by an edge are also called adjacent [be-
nachbart] (or neighbors).
Definition 5.2. The neighborhood of a vertex v is the set
Γ(v) := {u ∈ V | {u, v} ∈ E} of vertices adjacent to v. The
degree deg(v) of a vertex v is the number of edges (or ver-
tices) connected to v, i.e., deg(v) := |Γ(v)|. A graph is called
k-regular if deg(v) = k for all v ∈ V .
An important extension of the graph concept is obtained
when the edges are directed, i.e., they are ordered pairs (u, v)
(rather than sets {u, v}) where u and v are often called the
source and the destination, respectively, of the edge.
Definition 5.3. A directed graph G = (V, E) consists of a finite
set V of vertices and a set E ⊆ V × V of (directed) edges.
Definition 5.4. The in-degree deg− (v) of a vertex v is the num-
ber of edges entering v, and the out-degree deg+ (v) of v is the
number of edges leaving v.
P
Lemma 5.1. In a directed graph, v∈V deg− (v) = Definition 5.10. A graph G = (V, E) is contained in a graph
P + P
v∈V deg (v) = |E|. In an undirected graph, v∈V deg(v) =
2|E|.
17
The complete graph on n vertices, denoted Kn , is a simple
graph with n vertices in which any pair of vertices is con-
nected. The complement of Kn is the empty graph (with no
edges).
Definition 5.11. An (m, n)-mesh [Gittergraph] is a graph
Mm,n on mn vertices with V = {(i, j) | 1 ≤ i ≤ m, 1 ≤ j ≤
n} and where (i, j) and (i0 , j 0 ) are connected iff i = i0 and
|j − j 0 | = 1 or j = j 0 and |i − i0 | = 1.
Definition 5.12. A path [Pfad] Pn consists of n + 1 ver-
tices connected like a chain, i.e., V = {v0 , . . . , vn } and
E = {{v0 , v1 }, {v1 , v2 }, {v2 , v3 }, . . . , {vn−1 , vn }}. A directed
path P~n is like Pn , but it is directed, i.e., E = {(v0 , v1 ),
(v1 , v2 ), (v2 , v3 ), . . . , (vn−1 , vn )}.
Definition 5.13. A cycle [Kreis] Cn (for n ≥ 3) consists of n
vertices connected cyclically, i.e., V = {v1 , . . . , vn } and E =
{{v1 , v2 }, {v2 , v3 }, . . . , {vn , v1 }}. A directed cycle C~ n is like Cn ,
but it is directed, i.e., E = {(v1 , v2 ), (v2 , v3 ), . . . , (vn , v1 )}.
Definition 5.14. A d-dimensional hypercube Qd is a graph on
V = {0, 1}d with {u, v} ∈ E iff u and v differ in exactly one
bit.
Definition 5.15. The complete bipartite graph Km,n is a graph
on m + n vertices obtained by taking two vertex subsets B
and W (for black and white) of sizes m and n, respectively,
and connecting each vertex in B with every vertex in W , i.e.,
Km,n = (V, E) with V = B∪W , B∩W = ∅, |B| = m, |W | = n,
and E = {{u, v} | u ∈ B ∧ v ∈ W }.
19
Definition 5.5. A graph G = (V, E) is a subgraph [Teilgraph]
of a graph H = (V 0 , E 0 ), sometimes denoted G v H, if V ⊆ V 0
and E ⊆ E 0 .
Definition 5.6. The union of two graphs G = (V, E) and H =
(V 0 , E 0 ) is the graph G ∪ H := (V ∪ V 0 , E ∪ E 0 ).
The complement G of a graph G = (V, E) is the graph G =
(V, E) where E consist of all possible edges that are not in E.
Definition 5.7. A graph G = (V, E) is called bipartite if V can
be split into two disjoint sets V1 and V2 of vertices, V = V1 ∪V2 ,
such that no edge connects two vertices in the same subset Vi
(i = 1, 2).
Definition 5.8. The adjacency matrix AG = [aij ] of an undi-
rected graph G = (V, E) with
 V = {v1 , . . . , vn } is the binary
1 if {vi , vj } ∈ E
n × n matrix where ai,j =
0 otherwise.
For a directed graph, the condition {vi , vj } ∈ E must be re-
placed by (vi , vj ) ∈ E.
Definition 5.9. Two graphs G = (V, E) and H = (V 0 , E 0 ) are
isomorphic, denoted G ∼ = H, if there exists a bijection π : V →
V 0 such that renaming the vertices of G according to π results
in H, i.e., if {u, v} ∈ E ⇐⇒ {π(u), π(v)} ∈ E 0 .
For directed graphs the definition is similar, except that {u, v}
and {π(u), π(v)} must be replaced by (u, v) and (π(u), π(v)),
respectively.
H = (V 0 , E 0 ), denoted G  H, if there exists a subgraph K of
H that is isomorphic to G: G  H :⇔ ∃K (G ∼ = K ∧ K v H).
18
5.3 Paths and Cycles
Definition 5.16. A walk [Weg] (from u to v) of length n in
a graph or directed graph G is a sequence (u, v1 , . . . , vn−1 , v)
of vertices such that consecutive vertices are connected. If all
vertices are distinct, then a walk is called a path, and if all
the edges (but not necessarily the vertices) in the walk are dis-
tinct, it is called a tour [Tour]. When the starting and endpoint
are identical (i.e., u = v), then a path of length ≥ 3 is called a
cycle and a tour is called a circuit [Schleife].
Definition 5.17. An undirected graph G is connected [zusam-
menhängend] if any two vertices are connected by a path.
The maximal connected subgraphs of a graph G are called the
components of G.
Definition 5.18. A cycle in a (directed or undirected) graph
G is called Hamiltonian if it visits all vertices. If such a cycle
exists, the graph G is called Hamiltonian.
Theorem 5.2. A graph G = (V, E) for which |V | ≥ 3 and
deg(u)+deg(v) ≥ |V | for every non-adjacent pair (u, v) of vertices
(i.e., {u, v} ∈
/ E), is Hamiltonian. In particular, if deg(v) ≥ |V |/2
for all v ∈ V , then G is Hamiltonian.
Theorem 5.3. The hypercube Qd is Hamiltonian for d ≥ 2.
Definition 5.19. A Hamiltonian cycle in a hypercube is called
a Gray code.
20
5.4 Trees
Definition 5.24. A tree is an undirected connected graph with
no cycles. A forest is an undirected graph with no cycles. i.e.,
the union of several trees with disjoint vertex sets. A leaf is a
vertex with degree 1.
Lemma 5.6. A tree with n ≥ 2 vertices has at least 2 leaves.
Theorem 5.7. For a graph G with n vertices, the following state-
ments are equivalent:
(a) G is a tree.
(b) G has n − 1 edges and no cycles.
(c) G has n − 1 edges and is connected.
Definition 5.25. A spanning tree [Spannbaum] of a connected
graph G is a subgraph of G which is a tree and contains all
vertices of G.
Definition 5.26. A rooted tree is a tree with a distinguished
vertex, the root. There is a unique path from the root to every
vertex v; its length is the distance of v from the root. The
height or depth of the tree is the maximal distance of a leaf
from the root. The vertices on the path from the root to v are
called ancestors of v. The ancestor which is a neighbor of v is
called the parent, and v is called a child of the parent. A rooted
tree is a d-ary tree if every vertex has at most d children.
5.5 Planar Graphs
Definition 5.27. A graph is planar if it can be drawn in the
plane with no edges crossing.
21
Definition 5.30. A polyhedron is a solid bounded by a finite
number of (plane) polygon faces. The vertices and edges of
these polygons are the vertices and edges of the polyhedron.
A polyhedron is convex if the straight line segment connect-
ing any two points lies entirely within it. A polyhedron is
regular if for some m, n ≥ 3 each vertex meets exactly m faces
(and hence m edges) and each face is a regular n-gon.
Theorem 5.16. There are exactly five regular polyhedra, where
(m, n) is either (3, 3), (3, 4), (4, 3), (3, 5), or (5, 3).
23
Definition 5.28. A drawing of a planar graph divides the
plane into disjoint regions, one of which is infinite. The degree
of a region is the number of edges one encounters in a walk
around the region’s boundary. (An edge is counted twice if
the edge is a bridge.1 )
Theorem 5.8 (Euler’s formula). A plane drawing of a connected
plane graph G = (V, E) divides the plane into r := |E| − |V | + 2
regions.
Lemma 5.9. For any connected plane graph G = (V, E), the sum
of the degrees of the regions is equal to 2|E|.
Theorem 5.10. Every connected planar graph G = (V, E) with
|V | ≥ 3 satisfies |E| ≤ 3|V |−6. If G is bipartite, then the following
stronger inequality holds: |E| ≤ 2|V | − 4.
Corollary 5.11. Kn is planar iff n ≤ 4.
Corollary 5.12. K3,3 is not planar.
We can define three operations on a graph:
(1) deletion of edges,
(2) deletion of singleton vertices, and
(3) merging neighboring vertices, i.e., deleting the edge be-
tween them, replacing the two vertices by a single ver-
tex and maintaining all edges from the two (merged)
vertices.
Lemma 5.13. If a sequence of these three operations is performed
on a graph G and the resulting graph H is non-planar, then also G
is non-planar.
1 A bridge in a graph is an edge which, when removed, makes the graph
disconnected. A dangling edge is also a bridge.
22
6. Number Theory
6.2 Divisors and Division
Definition 6.1. For integers a and b with a 6= 0 we say that
a divides b, denoted a | b, if there exists an integer c such that
b = ac. In this case, a is called a divisor [Teiler] or factor
of b and b is called a multiple [Vielfaches] of a. The (unique)
integer c is called the quotient when b is divided by a, and we
write c = ab or c = b/a. We write a6 | b if a does not divide b.
Theorem 6.1 (Euclid). For all integers a and d 6= 0 there exist
unique integers q and r satisfying a = dq + r and 0 ≤ r < |d|.
Here a is called the dividend, d is called the divisor, q is called
the quotient, and r is called the remainder. The remainder r
is often denoted as Rd (a) or sometimes as a mod d.
Definition 6.2. For integers a and b (not both 0), an integer
d is called a greatest common divisor of a and b if d divides
both a and b and if every common divisor of a and b divides
d, i.e., d | a, d | b, and c | a ∧ c | b =⇒ c | d.
Definition 6.3. For a, b ∈ Z (not both 0) one denotes the
unique positive greatest common divisor by gcd(a, b) and
usually calls it the greatest common divisor. If gcd(a, b) = 1,
then a and b are called relatively prime [teilerfremd].
Definition 6.4. For a, b ∈ Z, the ideal generated by a and
b [durch a und b erzeugtes Ideal], denoted (a, b), is the set
(a, b) := {ua + vb | u, v ∈ Z}. Similarly, the ideal generated by
a single integer a is (a) := {ua | u ∈ Z}.
24
Lemma 6.2. For a, b ∈ Z there exists d ∈ Z such that (a, b) = (d). Lemma 6.6. If p is a prime which divides the product x1 x2 · · · xn
of some integers x1 , . . . , xn , then p divides one of them, i.e., p | xi
Lemma 6.3. Let a, b ∈ Z (not both 0). If (a, b) = (d), then d is a
for some i ∈ {1, . . . , n}.
greatest common divisor of a and b.
Corollary 6.4. For a, b ∈ Z (not both 0), there exist u, v ∈ Z such Theorem 6.7. Every positive integer can be written uniquely (up
that gcd(a, b) = ua + vb. to the order in which factors are listed) as the product of primes.
√
Theorem 6.8. n is irrational unless n is a square (n = c2 for
σ1 := a; σ2 := b; some c ∈ Z).
u1 := 1; u2 := 0;
v1 := 0; v2 := 1; Definition 6.6. The least common multiple l of two positive
while σ2 > 0 do begin integers a and b, denoted l = lcm(a, b), is the common multi-
q := σ1 div σ2 ; ple of a and b which divides every common multiple of a and
r := σ1 − qσ2 ; b, i.e., a | l, b | l, l > 0, and a | l0 ∧ b | l0 =⇒ l | l0 .
σ1 := σ2 ; σ2 := r;
t := u2 ; u2 := u1 − qu2 ; u1 := t;
t := v2 ; v2 := v1 − qv2 ; v1 := t; 6.5 Congruences and Modular Arithmetic
end;
Definition 6.8. For a, b, m ∈ Z with m ≥ 1, we say that a is
d := σ1 ; u = u1 ; v := v1 ;
congruent to b modulo m if m divides a − b. We write a ≡
b (mod m) or simply a ≡m b, i.e., a ≡m b :⇐⇒ m | (a − b).
Figure 6.1: Extended gcd-algorithm (Euclid).
Lemma 6.13. For any m ≥ 1, ≡m is an equivalence relation on Z.
Theorem 6.5. The algorithm described in Figure 6.1 computes, for Lemma 6.14. If a ≡m b and c ≡m d, then a + c ≡m b +
given nonnegative integers a and b with a ≥ b (not both 0), the d and ac ≡m bd.
integers d = gcd(a, b), as well as u and v satisfying ua + vb = Corollary 6.15. Let f (x1 , . . . , xk ) be a multi-variate polynomial
gcd(a, b). in k variables with integer coefficients, and let m ≥ 1. If ai ≡m bi
for 1 ≤ i ≤ k, then f (a1 , . . . , ak ) ≡m f (b1 , . . . , bk ).
6.3 Factorization into Primes
There are m equivalence classes of the equivalence relation
Definition 6.5. A positive integer p > 1 is called prime if the ≡m , namely [0], [1], . . . , [m − 1]. Each equivalence class [a]
only positive divisors of p are 1 and p. An integer greater than has a natural representative Rm (a) ∈ [a] in the set Zm :=
1 that is not a prime is called composite [zusammengesetzt]. {0, . . . , m − 1} of remainders modulo m.
25 26

Lemma 6.16. For any a, b, m ∈ Z with m ≥ 1, 7. Algebra

(i) a ≡m Rm (a).
(ii) a ≡m b ⇐⇒ Rm (a) = Rm (b).
Lemma 6.17. For any a, b, m ∈ Z with m ≥ 1,
(i) Rm (a + b) = Rm (Rm (a) + Rm (b)).
(ii) Rm (ab) = Rm (Rm (a) · Rm (b)).
Lemma 6.18. The congruence equation ax ≡m 1 has a solution
x ∈ Zm iff gcd(a, m) = 1. The solution is unique.
Definition 6.9. If gcd(a, m) = 1, the unique solution x ∈ Zm
to the congruence equation ax ≡m 1 is called the multi-
plicative inverse of a modulo m. One also uses the notation
x ≡m a−1 or x ≡m 1/a.
7.1 Introduction
Definition 7.1. An operation on a set S is a function S n → S,
where n ≥ 0 is called the “arity” of the operation.
Definition 7.2. An algebra (or algebraic system or Ω-algebra)
is a pair hS; Ωi where S is a set (the carrier [Trägermenge] of
the algebra) and Ω = (ω1 , . . . , ωn ) is a list of operations on S.1
7.2 Semigroups, Monoids, Groups
Definition 7.3. A left [right] neutral element (or identity el-
ement) of an algebra hS; ∗i is an element e ∈ S such that
e ∗ a = a [a ∗ e = a] for all a ∈ S. If e ∗ a = a ∗ e = a for
all a ∈ S, then e is simply called neutral element.

Theorem 6.20. Let m Lemma 7.1. If hS; ∗i has both a left and a right neutral element,
Q1r, m2 , . . . , mr be pairwise relatively prime then they are equal. In particular hS; ∗i can have at most one neu-
integers and let M = i=1 mi . For every list a1 , . . . , ar with 0 ≤
ai < mi for 1 ≤ i ≤ r, the system of congruence equations tral element.

x ≡m1 a1 Definition 7.4. A binary operation ∗ on a set S is associative

x ≡m2 a2 if a ∗ (b ∗ c) = (a ∗ b) ∗ c for all a, b, c, ∈ S.
... Definition 7.5. A semigroup [Halbgruppe] is an algebra hS; ∗i
x ≡mr ar where ∗ is associative.

for x has a unique solution x satisfying 0 ≤ x < M . Definition 7.6. A monoid is an algebra hM ; ∗, ei where ∗ is
associative and e is the neutral element.
Pr
This unique solution is x = RM ( i=1 ai Mi Ni ) with Mi = 1 This definition, though very general, does not capture all algebraic sys-

M/mi and Ni ≡mi Mi−1 . tems one might be interested in. A more general type of algebraic system,
called heterogeneous algebraic systems, can have several carrier sets.
27 28
Definition 7.7. A left [right] inverse element of an element
a in an algebra hS; ∗, ei with neutral element e is an element
b ∈ S such that b ∗ a = e [a ∗ b = e]. If b ∗ a = a ∗ b = e, then b
is simply called an inverse of a.
Lemma 7.2. In a monoid hM ; ∗, ei, if a ∈ M has a left and a right
inverse, then they are equal. In particular, a has at most one inverse.
7.3 Homomorphisms and Isomorphisms
Definition 7.10. For two compatible2 algebras hS; Ωi and
hS 0 ; Ω0 i, a function ψ : S → S 0 is called a homomorphism
from hS; Ωi to hS 0 ; Ω0 i if for every ω ∈ Ω (of arity n) and
corresponding ω 0 ∈ Ω0 (also of arity n), ψ (ω(a1 , . . . , an )) =
ω 0 (ψ(a1 ), . . . , ψ(an )) for every a1 , . . . , an ∈ S.

Definition 7.8. A group is an algebra hG; ∗i satisfying the fol- A mapping ψ from a group hG; ∗,b, ei to a group hG0 ; ?,b, e0 i is,
lowing axioms: by definition, a homomorphism if

1. ψ(e) = e0 ,
G1 ∗ is associative.
2. ψ(b d for all a, and
a) = ψ(a)
G2 There exists a (neutral) element e such that a∗e = e∗a =
a for all a ∈ G. 3. ψ(a ∗ b) = ψ(a) ? ψ(b) for all a and b.
G3 Every a ∈ G has an inverse element b
a, i.e., a ∗ b a∗a =
a=b Definition 7.11. A bijective homomorphism ψ from hS; Ωi to
e. hS 0 ; Ω0 i is called an isomorphism, and hS; Ωi and hS 0 ; Ω0 i are
called isomorphic, denoted hS; Ωi ∼ = hS 0 ; Ω0 i, if such an iso-
Definition 7.9. A group hG; ∗i (or monoid or semigroup) is morphism exists.
called commutative or abelian if a ∗ b = b ∗ a for all a, b ∈ G.

Lemma 7.3. For a group hG; ∗,b, ei, we have for all a, b, c ∈ G: 7.4 The Structure of Groups
c
(i) (b
a) = a. Definition 7.12. The direct product of n groups
hG1 ; ∗1 i , . . . , hGn ; ∗n i is the algebra
(ii) ad
∗ b = bb ∗ b
a. hG1 × · · · × Gn ; ?i, where the operation ? is component-wise:
(a1 , . . . , an ) ? (b1 , . . . , bn ) = (a1 ∗1 b1 , . . . , an ∗n bn ).
(iii) Left cancellation law: a ∗ b = a ∗ c ⇒ b = c.
Lemma 7.4. hG1 × · · · × Gn ; ?i is a group, where the neutral ele-
(iv) Right cancellation law: b ∗ a = c ∗ a ⇒ b = c.
ment and the inversion operation are component-wise in the respec-
(v) The equation a ∗ x = b has a unique solution x for any a and tive groups.
b. So does the equation x ∗ a = b. 2 Two algebras are compatible if there is a one-to-one correspondence be-

tween their operations, where corresponding operations have the same arity.
29 30

Definition 7.13. A subset H of a group hG; ∗,b, ei is called a Corollary 7.9. For a finite group G, the order of every elements
subgroup of G, denoted H ≤ G, if hH; ∗,b, ei is a group, i.e., if divides the group order, i.e., ord(a) divides |G| for every a ∈ G.
H is closed with respect to all operations:
Corollary 7.10. Let G be a finite group. Then a|G| = e for every
(1) a ∗ b ∈ H for all a, b ∈ H,
a ∈ G.
(2) e ∈ H, and
(3) b
a ∈ H for all a ∈ H. Corollary 7.11. Every group of prime order is cyclic, and in such
a group every element except the neutral element is a generator.
Definition 7.14. Let G be a group and let a be an element
of G. The order [Ordnung] of a, denoted ord(a), is the least
Definition 7.19. Z∗m := {a ∈ Zm | gcd(a, m) = 1}.
m ≥ 1 such that am = e, if such an m exists, and ord(a) = ∞
otherwise. Definition 7.20. The Euler function ϕ : Z+ → Z+ is defined
Definition 7.15. For a finite group G, |G| is called the order of as the cardinality of Z∗m : ϕ(m) = |Z∗m |.
G. Qr Qr
Lemma 7.12. if m = i=1 pei i , then ϕ(m) = i=1 (pi − 1)pei i −1 .
Lemma 7.5. In a finite group G, every element has a finite order.
If G is a group and a ∈ G has finite order, then for any m ∈ Z
we have am = aRord(a) (m) .
Definition 7.16. The smallest subgroup of a group G contain-
ing the element a ∈ G is called the group generated by a, de-
noted hai, is defined as hai := {an | n ∈ Z}.
Definition 7.17. A group G = hgi generated by an element
g ∈ G is called cyclic, and g is called a generator of G.
Theorem 7.6. A cyclic group of order n is isomorphic to hZn , ⊕i
(and hence abelian).
Theorem 7.8 (Lagrange). Let G be a finite group and let H be a
subgroup of G. Then the order of H divides the order of G, i.e., |H|
divides |G|.
31
Theorem 7.13. hZ∗m ; ,−1 , 1i is a group.
Corollary 7.14 (Fermat, Euler). For all m ≥ 2 and all a with
gcd(a, m) = 1, aϕ(m) ≡m 1. In particular, for every prime p and
every a not divisible by p, ap−1 ≡p 1.
Theorem 7.15. The group Z∗m is cyclic iff m = 2, m = 4, m = pe ,
or m = 2pe , where p is an odd prime and e ≥ 1.
Theorem 7.17. Let G be some finite group (multiplicatively writ-
ten), and let e ∈ Z be a given exponent relatively prime to |G| (i.e.
gcd(e, |G|) = 1). The (unique) e-th root of y ∈ G, namely x ∈ G
satisfying xe = y, can be computed according to x = y d , where d is
the multiplicative inverse of e modulo |G|, i.e., d ≡|G| e−1 .
32
7.5 Rings and Fields
Definition 7.21. A ring hR; +, −, 0, ·, 1i is an algebraic system
for which
(i) hR; +, −, 0i is an abelian group.
(ii) hR; ·, 1i is a monoid.
(iii) a(b+c) = ab+ac and (b+c)a = ba+ca for all a, b, c ∈ R
(left and right distributive laws).
A ring is called commutative if multiplication is commutative
(ab = ba).
Lemma 7.18. For any ring hR; +, −, 0, ·, 1i,
(i) 0a = a0 = 0 for all a ∈ R.
(ii) (−a)b = −ab.
(iii) (−a)(−b) = ab.
(iv) If R is non-trivial (i.e., if it has more than one element), then
1 6= 0.
Definition 7.22. The characteristic of a ring is the order of 1
in the additive group, but we define it to be 0 if the order of 1
is infinite.
Definition 7.23. Let R be a commutative ring. For a, b ∈ R
with a 6= 0 we say that a divides b, denoted a | b, if there exists
c ∈ R such that b = ac. In this case, a is called a divisor [Teiler]
or factor of b and b is called a multiple [Vielfaches] of a.
Lemma 7.19. In any commutative ring,
(i) If a | b and b | c, then a | c, i.e., the relation | is transitive.
(ii) If a | b, then a | bc for all c.
33
Definition 7.28. A field [Körper] is a nontrivial commutative
ring F in which every nonzero element is a unit, i.e., F ∗ =
F − {0}.
Theorem 7.24. Zp is a field iff p is prime.
Theorem 7.25. A field is an integral domain.
Theorem 7.26. A finite integral domain is a field.
7.6 Polynomials over a Field
Definition 7.29. A polynomial a(x) ∈ F [x] is called monic
[monisch, normiert] if the leading coefficient is 1.
Definition 7.30. A polynomial a(x) ∈ F [x] with degree at
least 1 is called irreducible if it is divisible only by constant
polynomials and by constant multiples of a(x).
Definition 7.31. For polynomials a(x) and b(x) in F [x] (not
both 0), a polynomial d(x) is called a greatest common divisor
of a(x) and b(x) if d(x) | a(x) and d(x) | b(x) and if every
common divisor of a(x) and b(x) divides d(x). Moreover, the
monic polynomial g(x) of largest degree such that g(x) | a(x)
and g(x) | b(x) is called the greatest common divisor of a(x)
and b(x), denoted gcd(a(x), b(x)).
Theorem 7.28. Let F be a field. For any a(x) and b(x) 6= 0 in F [x]
there exist unique q(x) (the quotient) and r(x) (the remainder) such
that a(x) = b(x) · q(x) + r(x) and deg(r(x)) < deg(b(x)).
Definition 7.35. Let a(x) ∈ R[x]. An element α ∈ R for which
a(α) = 0 is called a root [Nullstelle oder Wurzel] of a(x).
35
(iii) If a | b and a | c, then a | (b + c).
Definition 7.24. An element a 6= 0 of a commutative ring R is
called a zerodivisor [Nullteiler] if ab = 0 for some b 6= 0 in R.
Definition 7.25. An element u of a ring R is called a unit [Ein-
heit] if u is invertible, i.e., uv = vu = 1 for some v ∈ R (we
write v = u−1 ). The set of units of R is denoted by R∗ .
Lemma 7.20. For a ring R, R∗ is a multiplicative group (the group
of units of R).
Definition 7.26. An integral domain [Integritätsbereich] is a
nontrivial commutative ring without zerodivisors: ab = 0 ⇒
a = 0 ∨ b = 0.
Lemma 7.21. In an integral domain, if a | b, then c with b = ac is
unique (and is denoted by c = ab or c = b/a).3
Definition 7.27. A polynomial a(x) over a ring R in the in-
determinate x is a formal expression of the form a(x) =
Pd
ad xd + ad−1 xd−1 + · · · + a1 x + a0 = i=0 ai x . for some
i
non-negative integer d. The degree deg(a(x)) of a(x) is the
greatest i for which ai 6= 0. The special polynomial 0 (i.e., all
the ai are 0) is defined to have degree “minus infinity”. Let
R[x] denote the set of polynomials (in x) over R.
Theorem 7.22. R[x] is a ring.
Lemma 7.23. If D is an integral domain, then so is D[x]. The
units of D[x] are the constant polynomials which are units of D,
i.e., D[x]∗ = D∗ .
3 Note that the terms b
a
(or b/a) are defined only if a | b.
34
Lemma 7.31. For a field F , α ∈ F is a root of a(x) iff x−α divides
a(x).
Corollary 7.32. A polynomial a(x) of degree 2 or 3 over a field F
is irreducible iff it has no root.4
Definition 7.36. If α is a root of a(x), then its multiplicity is
the highest power of x − α dividing a(x).
Theorem 7.33. For an integral domain D, a nonzero polynomial
a(x) ∈ D[x] of degree d has at most d roots, counting multiplicities.
Lemma 7.34. A polynomial a(x) ∈ F [x] of degree d is uniquely
determined by any d + 1 values of a(x), i.e., by a(α1 ), . . . , a(αd+1 )
for any distinct α1 , . . . , αd+1 ∈ F .
Pd+1
This polynomial is then given by a(x) = i=1 βi ui (x), with
βi = a(αi ) and ui (x) = (α(x−α 1 )···(x−αi−1 )(x−αi+1 )···(x−αd+1 )
.
i −α1 )···(αi −αi−1 )(αi −αi+1 )···(αi −αd+1 )
7.7 Finite Fields
Let Rm(x) (a(x)) denote the (unique) remainder when a(x) is
divided by m(x). The concept of congruence modulo m(x)
is defined like congruence modulo m. For a(x), b(x) ∈ F [x],
a(x) ≡m(x) b(x) :⇐⇒ m(x) | (a(x) − b(x)).
Lemma 7.35. Congruence modulo m(x) is an equivalence relation
on F [x], and each equivalence class has a unique representative of
degree less than deg(m(x)).
Definition 7.37. Let m(x) be a polynomial of degree
d over F .
Then F [x]m(x) := a(x) ∈ F [x] | deg(a(x)) < d .
4 Note that this statement is not true for polynomials of degree ≥ 4.
36
Lemma 7.36. Let F be a finite field with q elements and let m(x)
be a polynomial of degree d over F . Then |F [x]m(x) | = q d .
Lemma 7.37. F [x]m(x) is a ring with respect to addition and mul-
tiplication modulo m(x).
Lemma 7.38. The congruence equation a(x)b(x) ≡m(x) 1 (for a
given a(x)) has a solution b(x) ∈ F [x]m(x) iff gcd(a(x), m(x))
 =
1. The solution is unique.5 In other words, F [x]∗m(x) = a(x) ∈
F [x]m(x) | gcd(a(x), m(x)) = 1 .
Theorem 7.39. The ring F [x]m(x) is a field iff m(x) is irreducible.6
7.8 Some Applications of Finite Fields
Definition 7.38. A (t, n)-secret sharing scheme for a finite do-
main S is a method for sharing a secret value s ∈ S among
n parties P1 , . . . , Pn such that any t of the parties can recon-
struct s, but no t − 1 (or fewer) parties have any information
about s.
Theorem 7.43. Let n < q and let each party Pi be (publicly)
assigned a unique element αi of GF(q). If a1 , . . . , at−1 are cho-
sen uniformly at random from GF(q) and each party Pi gets the
share a(αi ), where the polynomial a(x) ∈ GF(x) is defined by
a(x) := at−1 xt−1 +· · ·+a1 x+s, then this is a (t, n)-secret sharing
scheme.
Definition 7.39. The encoding function E of an error-
correcting code for some alphabet A takes k information
5 This b(x) (if it exists) is called the inverse of a(x) modulo m(x).
6 F [x] is called an extension field of F .
m(x)
37
8. Logic
8.1 Elementary Concepts in Logic
Let S be the set of (syntactic representations of) mathematical
statements. Every statement s ∈ S is either true or false. The
function τ : S → {0, 1} assigning to each s ∈ S its truth
value τ (s) and can be called the truth function. This function
defines the meaning, called the semantics, of objects in S.
Let P be the set of proofs, for example strings over some al-
phabet. A proof p for a statement s is relative to a verification
function φ : S ×P → {0, 1}, where φ(s, p) = 1 means that the
proof is accepted (by the function φ). The function φ defines
what is a proof for statement s.
Definition 8.1. A proof system is a quadruple Π
(S, P, τ, φ), as above.
Definition 8.2. A proof system Π = (S, P, τ, φ) is sound if no
false statement has a proof, i.e., if for all s ∈ S for which there
exists p ∈ P with φ(s, p) = 1, we have τ (s) = 1.
Definition 8.3. A proof system Π = (S, P, τ, φ) is complete
if every true statement has a proof, i.e., if for all s ∈ S with
τ (s) = 1, there exists p ∈ P with φ(s, p) = 1.
In addition, one requires that the function φ be efficiently
computable (for some notion of efficiency) and that every true
statement has a reasonably short proof.
Definition 8.4. The syntax of a logic defines an alphabet (of
allowed symbols) and specifies which strings (over the alpha-
bet) are (syntactically) correct formulas.
39
symbols a0 , . . . , ak−1 ∈ A and encodes them into a list
[c0 , . . . , cn−1 ] of n > k symbols in A (the codeword): E :
Ak → An : [a0 , . . . , ak−1 ] 7→ E(a0 , . . . , ak−1 ) = [c0 , . . . , cn−1 ].
Definition 7.40. An (n, k)-error-correcting code (or simply
(n, k)-code) C over the alphabet A with |A| = q is a subset
of cardinality q k of An .
Definition 7.41. The Hamming distance between two code-
words is the number of positions at which the two codewords
differ.
Definition 7.42. The minimum distance of an error-correcting
code C is the minimum of the Hamming distance between any
two codewords.
Definition 7.43. A decoding function D for an (n, k)-error-
correcting code is a function D : An → Ak .
Definition 7.44. A code C can correct t errors if there exists
a decoding function D such that when [r0 , . . . , rn−1 ] is ob-
tained from an arbitrary codeword [c0 , . . . , cn−1 ] by chang-
ing at most arbitrary t positions, then E (D([r0 , . . . , rn−1 ])) =
[c0 , . . . , cn−1 ].
Theorem 7.44. A code C with minimum distance d can correct t
errors iff d ≥ 2t + 1.
Theorem 7.45. Let A = GF(q) and let α0 , . . . , αn−1 be arbi-
trary distinct elements of GF(q). Consider the encoding function
E(a0 , . . . , ak−1 ) = [a(α0 ), . . . , a(αn−1 )], where a(x) is the poly-
nomial a(x) := ak−1 xk−1 +· · ·+a1 x+a0 . This code has minimum
distance n − k + 1.
38
Definition 8.5. A formula generally contains certain variable
parts which are not determined (by the formula) and can take
on values in certain domains. A particular choice of these
variable parts is called a structure.
Definition 8.6. A structure is suitable [passend] for a formula
F if all variable elements of F are defined (i.e., fixed), i.e., if it
makes the formula true or false.1
Definition 8.7. The semantics of a logic is a function σ as-
signing to each formula F and each structure A suitable for F
a truth value σ(F, A) in {0, 1}.
Definition 8.8. A (suitable) structure A for which a formula
F is true is called a model for F , and one writes A |= F. More
generally, for a set M of formulas, a (suitable) structure for
which all formulas in M are true is called a model for M , de-
=
noted as A |= M . If A is not a model for M one writes A 6|= M .
Definition 8.9. A formula F (or set M of formulas) is called
satisfiable [erfüllbar] if there exists a model for F (or M ), and
unsatisfiable otherwise. The symbol ⊥ is used for an unsatis-
fiable formula.
Definition 8.10. A formula F is called a tautology [Tautolo-
gie] or valid [gültig, allgemeingültig] if it is true for every suit-
able structure. The symbol > is used for a tautology.
Definition 8.11. A formula G is a logical consequence [(lo-
gische) Folgerung] of a formula F (or a set M of formulas),
denoted F |= G (or M |= G ), if every structure suitable for
both F (or M ) and G, which is a model for F (for M ), is also a
model for G.
1 A suitable structure can also define more variable elements then those
appearing in F .
40
Definition 8.12. Two formulas F and G are equivalent, de-
noted F ≡ G (or also F ⇐⇒ G), if every structure suitable for
both F and G yields the same truth value for F and G, i.e., if
each is logical consequence of the other: F ≡ G :⇐⇒ F |=
G and G |= F .
8.2 Logical Calculi
Definition 8.13. A derivation rule [Schlussregel] is a rule for
deriving a formula from a set of formulas (called the precon-
dition). We write {F1 , . . . , Fk } `R G if G can be derived
from the set {F1 , . . . , Fk } by rule R.
Definition 8.14. A (logical) calculus [Kalkül] K is a finite set
of derivation rules: K = {R1 , . . . , Rm }.
Definition 8.15. A derivation [Herleitung] of a formula G
from a set M of formulas in a calculus K is a finite sequence
(of some length n) of applications of rules in K, leading to G.
More precisely, we have M0 := M , Mi := Mi−1 ∪ {Gi } for
1 ≤ i ≤ n, where N `Ri Gi for some N ⊆ Mi−1 and for some
Ri ∈ K, and where Gn = G. We write M `K G if there is a
derivation of G from M in the calculus K.
Definition 8.16. A derivation rule R is correct if for every set
M of formulas and every formula F , M `R F =⇒ M |= F .
Definition 8.17. A calculus K is sound [widerspruchsfrei] or
correct if for every set M of formulas and every formula F , if
F can be derived from M then F is also a logical consequence
of M : M `K F =⇒ M |= F , and K is complete [vollständig]
if for every M and F , if F is a logical consequence of M , then
F can also be derived from M : M |= F =⇒ M `K F .
41
3) (F ∧G)∧H ≡ F ∧(G∧H) and (F ∨G)∨H ≡ F ∨(G∨H)
(associativity);
4) F ∧ (F ∨ G) ≡ F and F ∨ (F ∧ G) ≡ F (absorption);
5) F ∧ (G ∨ H) ≡ (F ∧ G) ∨ (F ∧ H) (distributive law);
6) F ∨ (G ∧ H) ≡ (F ∨ G) ∧ (F ∨ H) (distributive law);
7) ¬¬F ≡ F (double negation);
8) ¬(F ∧ G) ≡ ¬F ∨ ¬G and ¬(F ∨ G) ≡ ¬F ∧ ¬G
(de Morgan’s rules);
9) F ∨ > ≡ > and F ∧ > ≡ F (tautology rules);
10) F ∨ ⊥ ≡ F and F ∧ ⊥ ≡ ⊥ (unsatisfiability rules).
11) F ∨ ¬F ≡ > and F ∧ ¬F ≡ ⊥.
Definition 8.20. A literal is an atomic formula or the negation
of an atomic formula.
Definition 8.21. A formula F is in conjunctive normal form
(CNF) if it is a conjunction of disjunctions of literals, i.e., if it
is of the form F = (L11 ∨ · · · ∨ L1m1 ) ∧ · · · ∧ (Ln1 ∨ · · · ∨ Lnmn )
for some literals Lij .
Definition 8.22. A formula F is in disjunctive normal form
(DNF) if it is a disjunction of conjunctions of literals, i.e., if it
is of the form F = (L11 ∧ · · · ∧ L1m1 ) ∨ · · · ∨ (Ln1 ∧ · · · ∧ Lnmn )
for some literals Lij .
Theorem 8.3. Every formula is equivalent to a formula in CNF
and also to a formula in DNF.
Definition 8.23. A clause is a set of literals.
43
One writes `K F if F can be derived from the empty set of
formulas.
Lemma 8.1. If F `K G for a sound calculus, then |= (F → G).
8.3 Propositional Logic
Definition 8.18. (Syntax.) An atomic formula is of the form
Ai with i ∈ N.2 A formula is defined inductively: An atomic
formula is a formula, and if F and G are formulas, then also
¬F , (F ∧ G), and (F ∨ G) are formulas.
A formula of the form (F ∧ G) is called a conjunction, and a
formula of the form (F ∨ G) is called a disjunction.
Definition 8.19. (Semantics.) For a set M of atomic formulas,
a truth assignment [(Wahrheits-)Belegung] is a function A :
M → {0, 1}. Let M c be the set of formulas built from atomic
formulas in M . We extend the domain of A to M c, as follows:
A((F ∧ G)) = 1 if and only if A(F ) = 1 and A(G) = 1;
A((F ∨ G)) = 1 if and only if A(F ) = 1 or A(G) = 1;
A(¬F ) = 1 if and only if A(F ) = 0.
Lemma 8.2. For any formulas F , G, and H we have
1) F ∧ F ≡ F and F ∨ F ≡ F (idempotence);
2) F ∧ G ≡ G ∧ F and F ∨ G ≡ G ∨ F (commutativity);
2 A is usually not used. This definition guarantees an unbounded sup-
0
ply of atomic formulas, but as a notational convention we can also write
A, B, C, . . . instead of A1 , A2 , A3 , . . ..
42
Definition 8.24. The set of clauses associated to a for-
mula F = (L11 ∨ · · · ∨ L1m1 ) ∧ · · · ∧ (Ln1 ∨ · · · ∨
L
 nmn ) in CNF, denoted as K(F ), is the set K(F ) :=
{L11 , . . . , L1m1 } , . . . , {Ln1 , . . . , Lnmn } . The set of clauses
associated with a set M = {F1 , . . . , Fk } of formulas is the
Sk
union of their clause sets: K(M ) := i=1 K(Fi ).
Definition 8.25. A clause K is a resolvent of clauses K1 and
K2 if there is a literal L such that L ∈ K1 , ¬L ∈ K2 , and3
K = (K1 − {L}) ∪ (K2 − {¬L}). (8.1)
Given a set K of clauses, a resolution step takes two clauses
K1 ∈ K and K2 ∈ K, computes a resolvent K, and adds K to
K. To be consistent with Section 8.2.2, one can write the reso-
lution rule as {K1 , K2 } `res K, where equation (8.1) must
be satisfied.The resolution calculus, denoted Res, consists of a
single rule: Res = {res}.
Lemma 8.4. The resolution calculus is sound, i.e., if K `Res K
then K |= K.
Theorem 8.5. A set M of formulas is unsatisfiable iff
K(M ) `Res ∅.
3 For a literal L, ¬L is the negation of L, for example if L = ¬A, then
¬L = A.
44
8.4 Predicate Logic Definition 8.28. For a formula F , a variable x and a term t,
F [x/t] denotes the formula obtained from F by substituting
Definition 8.26. (Syntax of predicate logic.) every free occurrence of x by t.

• A variable is of the form xi with i ∈ N.4
(k)
• A function symbol is of the form fi with i, k ∈ N,
where k denotes the number of arguments of the func-
tion. Function symbols for k = 0 are called constants.
(k)
• A predicate symbol is of the form Pi with i, k ∈ N,
where k denotes the number of arguments of the predi-
cate.
• A term is defined inductively: A variable is a term, and
(k)
if t1 , . . . , tk are terms, then fi (t1 , . . . , tk ) is a term. For
k = 0 one writes no parentheses.
Definition 8.29. A structure is a tuple A = (U, φ, ψ, ξ) where
• U is a non-empty set, the so-called universe,
• φ is a function assigning to each function symbol (in a
certain subset of all function symbols) a function, where
for a k-ary function symbol f , φ(f ) is a function U k →
U.
• ψ is a function assigning to each predicate symbol (in
a certain subset of all predicate symbols) a function,
where for a k-ary function symbol P , ψ(P ) is a func-
tion U k → {0, 1}, and where

• A formula is defined inductively: • ξ is a function assigning to each variable symbol (in a

certain subset of all variable symbols) a value in U .
(k)
– If t1 , . . . , tk are terms, then Pi (t1 , . . . , tk ) is a for-
Definition 8.30. (Semantics.) For a structure A = (U, φ, ψ, ξ),
mula, called an atomic formula.
we define the value (in U ) of terms and the truth value of
– If F and G are formulas, then also ¬F , (F ∧ G), and formulas under that structure.
(F ∨ G) are formulas.
– ∀xi F and ∃xi F are also formulas. • The value A(t) of a term t is defined recursively as follows:

Definition 8.27. Every occurrence of a variable in a formula is – If t is a variable, then A(t) = ξ(t).
either bound or free. If a variable x occurs in a (sub-)formula – If t is of the form f (t1 , . . . , tk ) for terms t1 , . . . , tk
of the form ∀x G or ∃x G, then it is bound, otherwise it is free.5 and a k-ary function symbol f , then A(t) =
A formula is closed if it contains no free variables. φ(f )(A(t1 ), . . . , A(tk )).

• The truth value of a formula F is defined recursively as

4x is usually not used.
0 follows:
5 The occurrence of a variable x immediately following a quantifier is also

bound. – A((F ∧G)) = 1 if and only if A(F ) = 1 and A(G) = 1;

45 46

– A((F ∨ G)) = 1 if and only if A(F ) = 1 or A(G) = 1; Lemma 8.7. If one replaces a subformula G of a formula F by an
– A(¬F ) = 1 if and only if A(F ) = 0. equivalent (to G) formula H, then the resulting formula is equiva-
– If F is of the form F = P (t1 , . . . , tk ) for terms lent to F .
t1 , . . . , tk and a k-ary predicate symbol P , then Lemma 8.8. For a formula G in which x occurs only free and in
A(F ) = ψ(P )(A(t1 ), . . . , A(tk )). which y does not occur, ∀x G ≡ ∀y G[x/y] and ∃x G ≡
– If F is of the form ∀x G or ∃x G, then let A[x→u] be the ∃y G[x/y].
same structure as A except that ξ(x) is overwritten by
u (i.e., ξ(x) = u): By appropriately renaming quantified variables one can
 transform any formula into an equivalent formula in which
1 if A[x→u] (G) = 1 for all u ∈ U
A(∀x G) = no variable appears both as a bound and a free variable and
0 else
 such that all variables appearing after the quantifiers are dis-
1 if A[x→u] (G) = 1 for some u ∈ U tinct. Such a formula is said to be in rectified [bereinigt] form.
A(∃x G) =
0 else.
Definition 8.31. A formula of the form
Q1 x1 Q2 x2 · · · Qn xn G, where the Qi are arbitrary quantifiers
Lemma 8.6. Any equivalence that holds in propositional logic also (∀ or ∃) and G is a formula free of quantifiers, is said to be in
holds in predicate logic. Moreover, for any formulas F , G, and H, prenex form [Pränexform].
where H does not contain the variable x, we have


1) ¬(∀x F ) ≡ ∃x ¬F ; Theorem 8.9. ¬∃x∀y P (y, x) ↔ ¬P (y, y) .
2) ¬(∃x F ) ≡ ∀x ¬F ; Corollary 8.10. There exists no set that contains all sets S that do
3) (∀x F ) ∧ (∀x G) ≡ ∀x (F ∧ G); not contain themselves, i.e., {S| S 6∈ S} is not a set.
4) (∃x F ) ∨ (∃x G) ≡ ∃x (F ∨ G);
Corollary 8.11. The set {0, 1}∞ is not countable.
5) ∀x ∀y F ≡ ∀y ∀x F ;
6) ∃x ∃y F ≡ ∃y ∃x F ; Corollary 8.12. There are functions N → {0, 1} that are not com-
puted by any program.
7) (∀x F ) ∧ H ≡ ∀x (F ∧ H);
8) (∀x F ) ∨ H ≡ ∀x (F ∨ H);
9) (∃x F ) ∧ H ≡ ∃x (F ∧ H);
10) (∃x F ) ∨ H ≡ ∃x (F ∨ H).
47 48