MAA 201 2022-2023

Reduction of endomorphisms

Exercises - Week I - Groups

Exercise 1. 1. Show that (GLn (R), ∗) is a group.

2. Let U = {z ∈ C, |z| = 1}. Show that (U, ×) is a group.
3. Let E be a set. Show that the set of permutations of E endowed with
the composition law is a group, i.e. show that Bij(E) = ({f : E → E |
f is a bijection }, ◦) is a group.
 √
4. Show that G = x + y 3 | x ∈ N, y ∈ Z, x2 − 3y 2 = 1 is a subgroup of
(R∗+ , ×).

Solution:
1. Immediate from the definition of GLn (R).
2. U is the kernel of the morphism group z 7→ |z| from (C∗ , ×) to (R∗ , ×).
3. Composition of maps is associative. The neutral element is the identity, and bijective maps
of sets have an inverse.
√ √
4. Let g = x + y 3 be√in G. Then the inverse √ g is x − y 3 and is therefore in G.
0 0 0
Now, let g = x + y 3 and √ g = x 00+ y 03 be two elements of G.
We have gg 0 = √ x 00
+ y 00
3 with
√ x = xx + 3yy 0
and y 00 = xy 0 + x0 y.
First, x > |y| 3 and x0 > |y 0 | 3 shows √ that x
√
00
is positive. √
Then, √notice that √ a2 − 3b2 = (a + b 3)(a − b 3) for all a, b. And notice that x00 − y 00 3 =
(x − y 3)(x0 − y 0 3). Thus
√ √ √ √ √ √
(x00 + y 00 3)(x00 − y 00 3) = (x + y 3)(x0 + y 0 3)(x − y 3)(x0 − y 0 3)

and therefore, x002 − 3y 002 = 1.

Hence, G is stable by multiplication and inverse, it is indeed a subgroup of (R∗+ , ×).

Exercise 2.
Let G be a group. Is the intersection of two subgroups a group ? How about the
union ?

Solution: Yes the intersection of two subgroups if a subgroup. The union of two subgroups need
not be a subgroup. For example 2Z∪3Z contains 2 and 3 but not 2+3 = 5 so it is not a subgroup
of (Z, +).

Exercise 3.
Describe all possible group laws on a set G consisting of three elements.
 ∗ e a b
e e a b
Solution: By filling the following "sudoku" starting with one sees that the only
a a
b b
∗ e a b
e e a b
solution is
a a b e
b b e a
The reason why we cannot twice the same element in a row or a column is because, in a group,
we cannot have g ∗ x = g ∗ y if x 6= y.
Thus, any group with 3 elements is of the form {e, a, a2 } with a3 = e.

Exercise 4.
Let (E, ∗) be a set with a composition law ∗ which is associative, has a neutral
element, and such that every element has a left inverse. Show that every element
has a right inverse and that it coincides with its left inverse. Deduce from this that
E is a group.

Solution: Let h ∈ E and let g be it’s left inverse i.e. gh = e, we want to show that it is also h’s
right inverse. By hypothesis g has a left inverse x i.e. xg = e. We obtain x = x(gh) = (xg)h = h
which gives hg = e as desired.

Exercise 5.
Let (G, ∗) be a group.
1. Show that for any g ∈ G the map ϕg : G → G given by ϕg (x) = g ∗ x is a
bijection,
2. Show Φ : g 7→ ϕg is a homomorphism of groups from (G, ∗) to (Bij(G), ◦).
3. Check that Φ is injective.

Solution:
1. we have φg ◦ φg−1 (x) = g ∗ (g −1 x) = (gg −1 x) = x for all x ∈ G. Similarly φg−1 ◦ φg (x) = x.
Therefore φg is bijective with inverse φg−1 .
2. one checks as above that φg ◦ φh = φgh for all g, h ∈ G.
3. assume φg = φh , then we get with e the neutral element of G :

g = g ∗ e = φg (e) = φh (e) = h ∗ e = h

Therefore Φ is injective.

Exercise 6.
Let G be a group. An element g ∈ G is said to have (finite) order n > 1 if g n = e
but g m 6= e for all integers 1 6 m < n. If there is no such integer, we say that g
has infinite order. We assume here that G is finite.

Page 2
 1. Show that every element of G has finite order.
2. Assume the cardinality |G| of G (also called the order of G) is even. Show
that G has an element of order two.

Solution:
1. If g ∈ G has infinite order then g n 6= g m for all n 6=∈ Z. Indeed g n = g m would imply
g |n−m| = e. Therefore |G| = +∞.
2. Consider the set H = {g ∈ G | g −1 6= g}, it clearly has even order and cannot be equal
to G since e2 = e so there must exist a g 6= e which is not in H and which thus verifies
g 2 = e.

Exercise 7.
Let U be the subset of GL2 (R) defined by
  
1 a
U= , a∈R
0 1

1. Show U is an abelian subgroup of (GL2 (R), ∗).

2. Show U is isomorphic to (R, +).
 
1 a
Solution: Let Ua = . One easily checks that Ua Ub = Ua+b which means that a 7→ Ua is a
0 1
morphism of groups (clearly injective) from R to U. The image U is therefore a group isomorphic
to R.

Exercise 8.
Let G be a group and H be a subgroup of G. For g ∈ G we let gH := {gh, h ∈ H}.
1. Show that gH = H for g ∈ H.
2. Let g, g 0 ∈ G. Show that either gH = g 0 H or gH ∩ g 0 H = ∅.
3. Assume |G| < +∞. Prove that |H| divides |G| (you can firstly check that
|gH| = |H| for all g ∈ G).

Solution:
1. if g ∈ H, then as His a subgroup we have gh ∈ H for all h ∈ H, that is gH ⊂ H. To show
the converse inclusion, take h ∈ H. Then g −1 h ∈ H and therefore h = g(g −1 h) ∈ gH.
2. Assume gH ∩ g 0 H 6= ∅. There is h, h0 ∈ H such that gh = g 0 h0 . Therefore g −1 g 0 =
h(h0 )−1 ∈ H. By (1) we get g 0 H = g(g −1 g 0 H) = gH.
3. h 7→ gh is bijective from H to gH. Therefore |H| = |gH|. As G is finite we may find by
(2) a finite subset S of G such that G is the disjoint union of sH over s ∈ S. In particular
|G| = |S| × |H|.

Page 3
Exercise 9.
Translate in terms of group homomorphisms the following classical properties :
1. log(xy) = log(x) + log(y)
2. det(M N ) = det(M ) det(N )
3. |zz 0 | = |z||z 0 |
0 0
4. ez+z = ez ez
5. z + z 0 = z + z 0

Solution:
1. log is a group morphism from (R>0 , ×) to (R, +) (note that log(1) = 0).
2. det is a group morphism from GLn (k) to (k ∗ , ×) (note that det(Id) = 1).
3. | • | is a group morphism from (C∗ , ×) to (R>0 , ×) (note that |1| = 1).
4. exp is a group morphism from (R, +) to (R∗ , ×) (note that e0 = 1).
5. • is a group morphism from (C, +) to (C, +) (note that 0 = 0).

Exercise 10.
Describe all group homomorphisms from Z to Z. Which of them are injective and
which are surjective ?

Exercise 11.
Show that the groups
1. (R, +) and (R>0 , ×) are isomorphic
2. (R∗ , ×) and (C∗ , ×) are not isomorphic
3. (Q, +) and (Q>0 , ×) are not isomorphic

Solution:
1. exp gives an isomorphism whose inverse is log.
2. Suppose f : C∗ → R∗ is an isomorphism. We have f (−1)2 = f (1) = 1 thus f (−1) = ±1
but since f is bijective and f (1) = 1 we must have f (−1) = −1. Now we have f (i)2 =
f (−1) = −1 but this is impossible since there is no element in R∗ whose square is −1.
3. Suppose f : Q → Q>0 is an isomorphism. Then there exists√x ∈ Q such that f (x) = 2.
But x = x/2 + x/2 so 2 = f (x/2)2 which is impossible since 2 is not in Q>0 .

Page 4