Groups Flashcards
What is binary operation * in a set S?
A function
∗:S×S→S
(a,b)↦a∗b
When is a binary operation ∗ on a set S associative?
When a∗(b∗c) = (a∗b)∗c for all a,b,c∈S.
What is an identity element in a set S?
An element e∈S s.t e∗a = a = a∗e for all a∈S.
Let ∗ be a binary operation on a non-empty set S. Prove that if there is an identity e∈S, then it is unique.
Let e₁, e₂ be identity elements. Then e₁∗e₂ = e₂ as e₁ an identity and e₁∗e₂ = e₁ as e₂ an identity so e₁ = e₂.
Let ∗ be a binary operation on a set S, with identity e. Take a,b∈S, what will make b an inverse of a?
If a∗b = e = b∗a.
Let ∗ be an associative binary operation on a set S, with identity e. Take a∈S. Prove that if a has an inverse, then the inverse is unique.
Let b,b′ be inverses of a. Then b′∗(a∗b) = b′∗e = b′and (b′∗a)∗b = e∗b = b, but ∗ is associative so these are equal, so b = b′.
Let ∗ be a binary operation on a set S. Let T be a subset of S. We say that T is closed under ∗ if…
∗ : T×T→T is a binary operation on T.
A group is a set G together with a binary operation ∗ on G such that… [define the group axioms]
(i) ∗ is associative;
(ii) there is an identity;
(iii) each element of G has an inverse.
Let G be a group. Let g, g₁ ,g₂ ,g₃∈G, let m,n∈Z. Then prove:
(i) (g₁g₂)⁻¹ = g₂⁻¹g₁⁻¹;
(ii) (gⁿ)⁻¹ = (g⁻¹)ⁿ;
(iii) gᵐgⁿ = gᵐ⁺ⁿ;
(iv) (gᵐ)ⁿ=gᵐⁿ;
(v) if g₁g₂=g₁g₃, then g₂ = g₃ (‘cancellation on the left’);
(vi) if g₁g₂ = g₃g₂, then g₁ = g₃ (‘cancellation on the right’).
Proof left as an exercise to the flashcarder.
What makes a group (G,∗) Abelian?
If g∗g′ = g′∗g for all g,g′∈G — that is, if the binary operation ∗ is commutative.
What makes a group cyclic?
If there is some g∈G such that G = {gⁿ : n∈Z}.
Why is a cyclic group Abelian?
gⁿgᵐ = gⁿ⁺ᵐ = gᵐ⁺ⁿ = gᵐgⁿ
What is the nth cyclic group Cₙ?
{e,g,g²,…,gⁿ⁻¹}, where gⁿ=e, and for 0 ≤ i,j ≤ n−1 gᶦ∗gʲ =
{gᶦ⁺ʲ if i+j < n
{gᶦ⁺ʲ⁻ⁿ if i+j ≥ n
Let Pₙ be a regular n-gon in the plane (here n≥3). For n≥3, define the nth dihedral group D₂ₙ to be…
The set of isometries of the plane that send Pₙ to Pₙ. These isometries are called symmetries of Pₙ.
Let Pₙ be a regular n-gon in the plane. Write r for rotation anticlockwise by 2π/n about the centre of Pₙ, and s for reflection in an axis of Pₙ. Prove that the symmetries of Pₙ are e, r, r², . . . , rⁿ⁻¹, s, rs, r²s, . . . , rⁿ⁻¹s.
Label the vertices of Pₙ anticlockwise as 1, 2, . . . ,n. Let f be a symmetry of Pₙ, and consider f(Pₙ).
Case 1: The vertices of f(Pₙ) are numbered 1, 2, . . . , n anticlockwise. Say vertex 1 has moved to position k (where 1≤k≤n. Then applying(r⁻¹)ᵏ⁻¹ will return vertex 1 to position 1, and hence all vertices to their starting positions. So (r⁻¹)ᵏ⁻¹f=e, so f=rᵏ⁻¹.
Case 2: The vertices of f(Pₙ) are numbered 1, 2, . . . , n clockwise. Then fs keeps the vertices in anticlockwise order, so as in Case 1 we have fs = rᵏ⁻¹ for some k (1≤k≤n), and then f = fs² = rᵏ⁻¹s. So the symmetries of Pn are contained in the list e, r, r², . . . , rⁿ⁻¹, s, rs, r²s, . . . , rⁿ⁻¹s. Now e, r, r², . . . ,rⁿ⁻¹ each send vertex 1 to a different position, and hence are all distinct. Similarly, s, rs, r²s, . . . , rⁿ⁻¹s are all distinct. The former collection leave the vertices in anticlockwise order, whereas the latter switch them to clockwise, so in fact all 2n symmetries are distinct.
Given groups (G,∗𝓰) and (H,∗ₕ), we define their product group (or product) to be…
(G×H,∗) with ∗ defined by (g₁,h₁) ∗ (g₂,h₂) = (g₁ ∗𝓰 g₂, h₁ ∗ₕ h₂).
Prove that the operation * defined by (g₁,h₁) ∗ (g₂,h₂) = (g₁ ∗𝓰 g₂, h₁ ∗ₕ h₂) where (G,∗𝓰) and (H,∗ₕ) are groups makes (G×H,∗) into a group.
- closure: since ∗G and ∗H are binary operations on G and H respectively, we have (g₁ ∗𝓰 g₂, h₁ ∗ₕ h₂)∈G×H for all g₁, g₂∈G and h₁, h₂∈H.
- associativity: follows from the associativity of ∗𝓰 and ∗ₕ (exercise).
- identity: the identity element in (G×H, ∗) is e(𝓰 ₓ ₕ) = (e𝓰, eₕ).
- inverses: given (g, h)∈G×H, we have (g, h)⁻¹ = (g⁻¹,h⁻¹)∈G×H.
What is the order of a group (G,∗), and when is G a finite group?
The order is the cardinality |G| of the set G. If |G| is finite, then we say that G is a finite group.
Let G = {g₁,g₂,…,gₙ} be a finite group. The Cayley table, or group table, of G is…
A square n×n grid in which the entry in row i and column j is gᶦ∗gʲ.
Prove that a Cayley table is a Latin square: each element appears exactly once in each row and in each column.
Let G be a finite group, take g∈G. Define f𝓰 : G→G g′↦gg′. This is a bijection (its inverse is g′↦g⁻¹g′).
So the entries in the row corresponding to g are precisely the elements of G in some order, each appearing exactly once.
Similarly for columns.
Let (G,∗) be a group. We say that a subset H⊆G is a subgroup (H≤G) if…
The restriction of ∗ to H makes H into a group, that is,
• H is closed under ∗ (if h₁,h₂∈H then h₁h₂∈H);
• H has an identity (e𝓰∈H);
• H contains inverses (if h∈H then h⁻¹∈H).
Let G be a group, and take g∈G. We define the order of g, o(g), to be…
When does g have infinite order?
the smallest positive integer k such that gᵏ = e. If no such integer k exists, then we say that g has infinite order.
Let (G,∗𝓰) and (H,∗ₕ) be groups. An isomorphism between G and H is…
A bijective map θ : G→H such that θ (g₁ ∗𝓰 g₂) = θ(g₁) ∗ₕ θ(g₂) for all g₁,g₂∈G. If such an isomorphism exists, then we say that G and H are isomorphic, and write G∼=H.
What is a permutation of a set S?
A bijection S→S
What is the set of permutations of a set S called?
Sym(S)
What is shorthand for τ(σ(k))?
kστ
Let S be a set. Prove that
(i) Sym(S) is a group under composition, called the symmetry group of S.
(ii) If |S|≥3, then Sym(S) is non-Abelian.
(iii) |Sₙ|=n!.
(i) Exercise. (ii) Let x₁, x₂, x₃ be three distinct elements of S. We seek to find two elements of Sym(S) that don’t commute. Define f: x₁↦x₂ x₂↦x₁ x↦x for other x and g: x₂↦x₃ x₃↦x₂ x↦x for other x. Then f,g∈Sym(S) and x₁gf = x₁f = x₂ while x₁fg = x₂g = x₃ so fg =/= gf.
(iii) To specify f∈Sₙ, we say what f does to each of 1, 2, … , n. There are n possibilities for 1f, and f is injective so there are n−1 possibilities for 2f, and so on.This gives n! possibilities for f.
A permutation σ∈Sₙ is a cycle if there are distinct a₁, . . . ,aₖ∈{1,2,…,n} such that…
What is the name for this k-length cycle and how is it wriiten?
aᵢσ = aᵢ₊₁ for 1≤i≤k−1 and aₖσ = a₁ and xσ = x for x ∈/ {a₁,…,aₖ}. We call it a k-cycle, and this has order k. We write it as (a₁ a₂ … aₖ).
The cycles (a₁… aₖ) and (b₁… bₗ) are disjoint if…
aᵢ =/= bⱼ for all i, j.
Let α = (a₁… aₖ) and β = (b₁… bₗ) be disjoint cycles. Show that α and β commute.
We have
aᵢαβ = aᵢ₊₁β = aᵢ₊₁ for 1≤i≤k−1 and aᵢβα = aᵢα = aᵢ₊₁ for 1≤i≤k−1 and aₖαβ = a₁β = a₁ and aₖβα = aₖα = a₁. Similarly bⱼαβ = bⱼβα for 1≤j≤l, and for x ∈/ {a₁,…,aₖ,b₁,…,bₗ} we have xαβ = x = xβα. So αβ = βα.
Show that every permutation in Sₙ can be written as a product of disjoint cycles. And that this product is unique up to cycling elements without cycles and permuting the order of the cycles.
Bottom of page 13.
[Two parts, existence and uniqueness]
What is the cycle type for a given permutation σ∈Sₙ?
The list of lengths of the cycles in σ, e.g The permutation (1 5 3 9)(2 6)(4 7)∈S₉ has cycle type 4, 2, 2, 1.