Introduction to Topological Manifolds, John M. Lee, Second Edition
Appendix A: Review of Set Theory
Basic Concepts
▶ Exercise A.1 Suppose A is a set and C is a collection of sets. Prove the following properties of unions and intersections.
(a) Distributive Laws:
A∪(X∈C⋂X)=X∈C⋂(A∪X);
A∩(X∈C⋃X)=X∈C⋃(A∩X).
Solution
Let x∈A∪(⋂X∈CX). If x∈A then clearly we have x∈⋂X∈C(A∪X) because x∈A⟹x∈A∪X for all X∈C. Alternatively if x∈⋂X∈CX we also have x∈⋂X∈C(A∪X) because for each X∈C we have x∈X⊆A∪X. Therefore A∪(⋂X∈CX)⊆⋂X∈C(A∪X).
Let x∈⋂X∈C(A∪X). We propose that x∈A∪(⋂X∈CX). If x∈A than this is trivial. If x∈/A, then we can use the fact that for each X∈C we have x∈A∪X to conclude that x∈X for each X∈C. This can be rewritten as x∈⋂X∈CX. This implies that the proposition is true. Therefore ⋂X∈C(A∪X)⊆A∪(⋂X∈CX).
Let x∈A∩(⋃X∈CX). This means that x∈A and that for some X′∈C we have x∈X′. Therefore x∈A∩X′, so x∈⋃X∈C(A∩X). Since x is arbitrary we have A∩(⋃X∈CX)⊆⋃X∈C(A∩X).
Let x∈⋃X∈C(A∩X). Then for some X′∈C we have x∈A∩X′. Since we have x∈A and x∈⋃X∈CX (since X′⊆⋃X∈CX) we have x∈A∩(⋃X∈CX). Therefore ⋃X∈C(A∩X)⊆A∩(⋃X∈CX).
Combining the four subset relations at the end of each paragraph gives the two desired set equalities. □
(b) De Morgan’s Laws:
A∖(X∈C⋂X)=X∈C⋃(A∖X);
A∖(X∈C⋃X)=X∈C⋂(A∖X).
Solution
Let x∈A∖(⋂X∈CX). Then there must exist some X′∈C such that x∈/X′, because if there wasn't we would have x∈⋂X∈CX, contradicting our choice of x. Since x∈A as well we have x∈A∖X′. Since X′∈C we have x∈⋃X∈C(A∖X). Since our choice of x was arbitrary we have A∖(⋂X∈CX)⊆⋃X∈C(A∖X).
Let x∈⋃X∈C(A∖X). Then there exists some X′∈C such that x∈A∖X′. So x∈A and x∈/X′. Since x∈/X′ we must have x∈/⋂X∈CX because x∈⋂X∈C⟹x∈X′. Since x∈A and x∈/⋂X∈CX we have x∈A∖(⋂X∈CX). Since our choice of x was arbitrary we have ⋃X∈C(A∖X)⊆A∖(⋂X∈CX).
Let x∈A∖(⋃X∈CX). Let X′∈C. By our choice of x we have x∈A and x∈/⋃X∈CX. We must have x∈/X′, because x∈X′⟹x∈⋃X∈CX. So we have x∈A∖X′ for each X′∈C as our choice of X′ was arbitrary. We can rewrite this as x∈⋂X∈C(A∖X). Since our choice of x was arbitrary we have A∖(⋃X∈CX)⊆⋂X∈C(A∖X).
Let x∈⋂X∈C(A∖X). Clearly x∈A. Let X∈C. By our choice of x we have x∈A∖X, so x∈/X. Since this applies to all X∈C, we have x∈/⋃X∈CX. So x∈A∖(⋂X∈CX). Since our choice of x was arbitrary, we have ⋂X∈C(A∖X)⊆A∖(⋂X∈CX).
Combining the four subset relations at the end of each paragraph gives the two desired set equalities. □
Cartesian Products, Relations, and Functions
Relations
▶ Exercise A.2. Given an equivalence relation ∼ on a set X, show that the set X/∼ of equivalence classes is a partition of X. Conversely, given a partition of X, show that there is a unique equivalence relation whose set of equivalence classes is exactly the original partition.
Solution
First, we need to show that X/∼ is a partition of X. By the definition of a partition we need to show that X is the disjoint union of the sets in X/∼. By the definition of a disjoint union we need to show that A,B∈X/∼ and A=B⟹A∩B=∅ (this is the "disjoint") and that {y∈[x]:x∈X}=X (this is the "union").
Let A,B∈X/∼ such that A=B. Let a,b∈X such that A=[a], B=[b]. Suppose there exists some c∈A∩B. Since a∼c and b∼c we have a∼b⟹A=B, a contradiction. By contradiction there exists no c∈A∩B, so we conclude that A∩B=∅, the "disjoint" part of the "disjoint union".
Let y∈{y∈[x]:x∈X}, then for some x∈X we have y∈[x]⟹y∼x⟹y∈X, so {y∈[x]:x∈X}⊆X. Let x∈X, then x∈[x], so X⊆{y∈[x]:x∈X}. Therefore {y∈[x]:x∈X}=X, the 'union' part of the "disjoint union".
Conversely, let C be a partition of X. Let ∼ be a relation on X such that a∼b⟺∃U∈C s.t. a,b∈U. We need to show that ∼ is an equivalence relation. Let a,b,c∈X. Since X is the disjoint union of the sets of C, there must exist some U∈C such that a∈U, so a∼a. It is trivial that a∼b⟹b∼a. If a∼b and b∼c, then there must exist U,V∈C such that a,b∈U and b,c∈V. Since U∩V=∅ we must have U=V (this is the contrapositive of the disjoint rule), so a∼c. Therefore ∼ is an equivalence relation as desired. □
▶ Exercise A.3. Let R⊆X×X be any relation on X, and define ∼ to be the intersection of all equivalence relations in X×X that contain R.
(a) Show that ∼ is an equivalence relation.
Solution
Let x,y,z∈X. First we show that x∼x. This is obvious because, for each equivalence relation ∼′⊇R, we have (x,x)∈∼′, so we have (x,x)∈⋂∼′⊇R,∼′an equiv relation∼′=∼. The same argument works for showing symmetry and trasitivity.
(b) Show that x∼y if and only if at least one of the following statements is true: x=y, or xR′y, or there is a finite sequence of elements z1,…,zn∈X such that xR′z1R′⋯R′znR′y, where xR′y means "xRy or yRx."
Solution
Assume x≁y, so there is some equivalence relation ∼′⊇R such that x≁′y. We can not have x=y, because it would contadict x∼′x (since ∼′ is an equivalence relation). Since ∼′⊇R, we have xR′y⟹xRy or yRx⟹x∼′y or y∼′x⟹x∼′y, so we can not have xR′y. Similarly, xR′z1R′⋯R′znR′y⟹x∼′z1∼′⋯∼′zn∼′y⟹x∼′y, so we can not have xR′z1R′⋯R′znR′y.
Alternatively, assume that none of the three conditions are true. We need to show that x≁y, and we can do this by constructing an equivalence relation ∼′⊇R such that x≁′y. Use the three conditions to construct the equivalence relation. This construction is indeed an equivalence relation (does this need a quick verification?) and does not contain (x,y). □
Functions
▶ Exercise A.4. Let f:X→Y and g:W→X be maps, and suppose R⊆W,S,S′⊆X, and T,T′⊆Y. Prove the following:
(a)T⊇f(f−1(T)).
(b)T⊆T′⟹f−1(T)⊆f−1(T′).
(c)f−1(T∪T′)=f−1(T)∪f−1(T′).
(d)f−1(T∩T′)=f−1(T)∩f−1(T′).
(e)f−1(T∖T′)=f−1(T)∖f−1(T′).
(f)S⊆f−1(f(S)).
(g)S⊆S′⟹f(S)⊆f(S′).
(h)f(S∪S′)=f(S)∪f(S′).
(i)f(S∩S′)⊆f(S)∩f(S′).
(j)f(S∖S′)⊇f(S)∖f(S′).
(k)f(S)∩T=f(S∩f−1(T)).
(l)f(S)∪T⊇f(S∪f−1(T)).
(m)S∩f−1(T)⊆f−1(f(S)∩T).
(n)S∪f−1(T)⊆f−1(f(S)∪T).
(o)(f∘g)−1(T)=g−1(f−1(T)).
(p)(f∘g)(R)=f(g(R)).
▶ Exercise A.5. With notation as in the previous exercise, give counterexamples to show that the following equalities do not necessarily hold true.
(a)T=f(f−1(T)).
(b)S=f−1(f(S)).
(c)f(S∩S′)=f(S)∩f(S′).
(d)f(S∖S′)=f(S)∖f(S′).
▶ Exercise A.6. Show that a composition of injective functions is injective, a composition of surjective functions is surjective, and a composition of bijective functions is bijective.
▶ Exercise A.7. Show that equality (a) in Exercise A.5. holds for every T⊆Y if and only if f is surjective, and each of the equalities (b)—(d) holds for every S,S′⊆X if and only if f is injective.
▶ Exercise A.8. Let f:X→Y be a function.
(a) Show that f has an inverse if and only if it is bijective.
(b) Show that if f has an inverse, its inverse is unique.
(c) Show that if f:X→Y and g:Y→Z are both bijective, then (g∘f)−1=f−1∘g−1.
▶ Exercise A.10. Show that if f:X→Y is bijective, then any left or right inverse for f is equal to f−1.
Number Systems and Cardinality
▶ Exercise A.11. Show that the real numbers are unique, in the sense that any complete ordered field admits a bijection with R that preserves addition, multiplication, and order.
▶ Exercise A.12. Show that an interval must be one of the nine types of sets [a,b], (a,b), [a,b), (a,b], (−∞,b], (−∞,b), [a,∞), (a,∞), or (−∞,∞).
▶ Exercise A.13. Prove that any susbset of a countable set is countable.
▶ Exercise A.14. Prove that the Cartesian product of two countable sets is countable.
Indexed Families
▶ Exercise A.15. Complete the proof of Lemma A.9 by showing that every surjective function has a right inverse.
▶ Exercise A.16. Prove that if there exists a surjective map from a countable set onto S, then S is countable.
▶ Exercise A.17. Prove that the union of a countable collection of countable sets is countable.