This document consolidates all proofs you should be able to reproduce under exam conditions, organised by topic.
Part I: Proofs
1. Irrationality of 2
Theorem: 2 is irrational.
Proof:
Suppose for contradiction 2=qp where gcd(p,q)=1.
Squaring: 2=q2p2, so p2=2q2.
Thus p2 is even, so p is even. Write p=2k.
Substituting: 4k2=2q2, so q2=2k2.
Thus q2 is even, so q is even.
But p and q both even contradicts gcd(p,q)=1. □
2. Infinitude of Primes (Euclid)
Theorem: There are infinitely many primes.
Proof:
Suppose finitely many primes: p1,…,pn.
Let N=p1⋯pn+1.
Since N>1, N has a prime divisor p.
If p=pi for some i, then p∣p1⋯pn and p∣N, so p∣(N−p1⋯pn)=1.
No prime divides 1. Contradiction.
Thus p is not in our list. □
3. Fermat’s Little Theorem
Theorem: If p is prime and p∤a, then ap−1≡1(modp).
Proof:
Consider a,2a,…,(p−1)a(modp).
Non-zeroness: Each ia≡0(modp) since p∤a and p∤i.
Distinctness: If ia≡ja(modp), then p∣(i−j)a. By Euclid’s lemma, p∣(i−j). But ∣i−j∣<p, so i=j.
Thus {a,2a,…,(p−1)a} = {1,2,…,p−1}(modp).
Taking products: ap−1(p−1)!≡(p−1)!(modp).
Since gcd((p−1)!,p)=1, cancel: ap−1≡1(modp). □
Part II: Sets, Relations, Functions
4. Relational Composition is Associative
Theorem: (T∘S)∘R=T∘(S∘R).
Proof:
a((T∘S)∘R)d⟺∃c.aRc∧c(T∘S)d
⟺∃c,b.aRc∧cSb∧bTd
a(T∘(S∘R))d⟺∃b.a(S∘R)b∧bTd
⟺∃b,c.aRc∧cSb∧bTd
Both equivalent to ∃b,c.aRc∧cSb∧bTd. □
5. Identity Laws for Composition
Theorem: idB∘R=R=R∘idA.
Proof:
a(idB∘R)b⟺∃c.aRc∧cidBb⟺∃c.aRc∧c=b⟺aRb.
a(R∘idA)b⟺∃c.aidAc∧cRb⟺∃c.a=c∧cRb⟺aRb. □
6. Reflexive-Transitive Closure is Smallest Preorder
Theorem: R∘∗ is the smallest preorder containing R.
Proof:
R∘∗ is a preorder: reflexive (contains id), transitive (path concatenation).
R⊆R∘∗: Clear since R∘1=R.
Smallest: Let P be a preorder with R⊆P. By induction on n: R∘n⊆P.
- Base: id⊆P by reflexivity.
- Step: R∘(n+1)=R∘R∘n⊆P∘P⊆P by transitivity.
Thus R∘∗=⋃nR∘n⊆P. □
7. Equivalence Relations ↔ Partitions Bijection
Theorem: Φ:EqRel(A)→Part(A) by Φ(E)={[a]E:a∈A} is a bijection.
Proof:
Well-defined: Equivalence classes partition A (non-empty by reflexivity, disjoint by transitivity/symmetry, cover by definition).
Inverse: Given partition P, define E={(a,b):∃B∈P.a,b∈B}.
This is an equivalence relation (reflexive: a in unique block; symmetric: trivial; transitive: blocks disjoint).
Inverses: Check Φ−1(Φ(E))=E (elements in same class iff equivalent) and Φ(Φ−1(P))=P (blocks are the equivalence classes). □
8. Isomorphisms are Bijective
Theorem: f:A→B is an isomorphism iff f is bijective.
Proof:
(⇒): If f−1 exists: injective (if f(a1)=f(a2), then a1=f−1(f(a1))=f−1(f(a2))=a2); surjective (for b, b=f(f−1(b))).
(⇐): Define f−1(b)= the unique a with f(a)=b. Verify f−1∘f=id and f∘f−1=id. □
9. Sections and Idempotents
Theorem: (s,r) section-retraction pair ↔ e=s∘r idempotent.
Proof:
(⇒): e∘e=s∘r∘s∘r=s∘id∘r=s∘r=e.
(⇐): Given idempotent e, let B=Im(e). Define r(a)=e(a), s(b)=b (inclusion). Then r(s(b))=e(b)=b since b∈Im(e) means b=e(a) for some a. □
10. Cantor’s Theorem
Theorem: No surjection A↠P(A) exists.
Proof:
Suppose e:A↠P(A). Define U={x∈A:x∈/e(x)}.
Since e surjective, ∃a.e(a)=U.
Is a∈U?
- If yes: a∈U⇒a∈/e(a)=U. Contradiction.
- If no: a∈/U⇒a∈e(a)=U. Contradiction.
Thus no such e exists. □
11. Well-Founded Induction
Theorem: If (∀y≺x.P(y))⇒P(x) for all x, then ∀x.P(x).
Proof:
Suppose S={x:¬P(x)}=∅.
By well-foundedness, S has minimal element m.
For all y≺m: y∈/S, so P(y) holds.
By assumption, P(m) holds, contradicting m∈S.
Thus S=∅. □
12. Rule Induction
Theorem: If S is closed under all rules in R, then XR⊆S.
Proof:
Induction on derivation height.
Base: Axioms have height 1. Since S closed under rules with no premises, axioms are in S.
Step: Derivation of height n+1 uses rule vu1,…,uk where each ui derived by height ≤n.
By IH, each ui∈S. Since S closed under the rule, v∈S. □
13. Subset Construction Correctness
Theorem: L(P(M))=L(M).
Proof:
Soundness (L(P(M))⊆L(M)): DFA path s′→Sn∩F=∅. Build NFA path by tracing elements in each Si back to NFA states. Some qf∈Sn∩F reachable from s.
Completeness (L(M)⊆L(P(M))): NFA path s⇒wqf. Track reachable states Si at each step. qf∈Sn, so Sn∩F=∅. □
14. Kleene’s Theorem
Theorem: L is regular iff L=L(R) for some regex R.
Proof:
Regex ⇒ NFA: Inductive construction (Thompson) for ε, ∅, symbols, union, concat, star. Each preserves language.
NFA ⇒ Regex: Region decomposition. Compute R(q,S,q′) for paths through regions. Eliminate all states: L(M)=⋃qf∈FR(s,Q,qf). □
15. Pumping Lemma
Theorem: For L regular, ∃p.∀w∈L with ∣w∣≥p: w=xyz with ∣xy∣≤p, ∣y∣≥1, and ∀i.xyiz∈L.
Proof:
Let DFA have n states. Set p=n.
For ∣w∣≥n: computation visits n+1 states. By pigeonhole, some state repeated.
Let x = prefix before loop, y = loop, z = suffix.
- ∣xy∣≤n=p (loop within first n states)
- ∣y∣≥1 (loop non-trivial)
- Can repeat or omit loop; stays accepting. □
Summary
These 15 proofs form the core of examinable material. Key strategies:
- Contradiction: Assume negation, derive 2, primes, Cantor
- Induction: Natural numbers, derivation height, state elimination
- Construction: Explicit bijections, automata constructions
- Unfolding definitions: Relational composition, closure properties