9/10/22, 3:51 AM 3111 Group Theory

Art of Problem Solving

Group Theory (3111)

Christian Hokaj

Friday
Jun 17, 2022 - Sep 16, 2022
7:30 - 9:30 PM ET (4:30 - 6:30 PM PT)

Homework

Week: 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Homework: Week 8
You have completed 12 of 12 challenge problems.
You did not submit a writing problem response.
Past Due Aug 13.
Request Extension

Readings
Week 8: Chapter 7
Week 9: Chapter 8
Week 8 Transcript: p Fri, Aug 5

Challenge Problems

Problem 1 – Give Up – Score: 0 / 7 (38755)

Problem: Report Error

Which coset of , viewed as a subgroup of , contains the permutation ?

(15)S4 (25)S4 (35)S4 (45)S4

Solution:
If , the permutation takes to and to . Thus can only be in . Indeed, we have
.

Your Response(s):
& Give Up

Problem 2 – Give Up – Score: 0 / 7 (38796)

Problem: Report Error

Identify the isomorphism class of the group

C
6

Solution:
Name our group . We begin by focusing on the third relation , that is, . This implies
. Squaring both sides shows , so that in fact . Now our
relation simplified to , so is Abelian!

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 1/9
9/10/22, 3:51 AM 3111 Group Theory

To summarize so far, we've rewritten our group as . Now since is Abelian, we have
. If then , but this is not implied by any of our relations unless is divisible by .
Thus we have . Finally we note that and , so we can generate by
, with order . Thus we find

Your Response(s):
& Give Up

Problem 3 – Give Up – Score: 0 / 7 (38947)

Problem: Report Error

How many vertices in the Cayley graph of can be reached within steps, starting at the identity?

⎛ ⎞

⎜ ⎟
+ ⎜ − ⎟
⎜ ⎟

⎝ ⎠

1 2 3 4 n n − 1 n + 1

Solution:
We can reach vertex in steps and vertices within step. After that, with each further step, we can reach three new
vertices from each vertex that was new in the previous step.

Therefore, let's define to be the number of vertices we can reach in steps but no fewer. We have and
for we established above that Therefore is a geometric sequence, with for

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 2/9
9/10/22, 3:51 AM 3111 Group Theory

Now, the number of vertices we can reach in or fewer steps is

In particular, we see that the number of vertices reachable in steps on this Cayley graph grows exponentially. This is in sharp
contrast to the polynomial growth of the Cayley graph of seen in last week's homework; this distinction indicates
quantitatively that "has a lot of relations between words in the generators," in a sense.

Your Response(s):
& Give Up

Problem 4 – Correct! – Score: 5 / 7 (38744)

Problem: Report Error

We can view as a subgroup of , since any isometry of the square is given by a permutation of the vertices. Is it a normal
subgroup?

Yes No

Solution:
Solution 1: Write If were to be
normal in we would need to have for any and However, we know that all
permutations of the same cycle type are conjugate in Since does not contain the whole conjugacy class formed by the
-cycles, nor that formed by the transpositions, it is

Solution 2: Write If were

normal in then the set of left cosets would be a group under set multiplication. Then a generator should be some
coset such that but This implies and since no element of has
order or we must have in This helps motivate consideration of the coset
If is to be normal, we must
have However, this is not the case; for instance is not in
Therefore is not normal in

Your Response(s):
% Yes
$ No

Problem 5, Part (a) – Give Up – Score: 0 / 7 (38806)

Problem: Report Error

Consider the group with presentation What is 's order?

Solution:
The relation implies . Thus we can move the generator past the generator, to write every
element as a product of s and s. We have which means . Furthermore, we can simplify to
and to

Thus we see that every element of may be written in the form with and . If this

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 3/9
9/10/22, 3:51 AM 3111 Group Theory

expression can be shown to be unique, then we'll be done. Now, if , we have . The only
relation that allows us to rewrite a power of as a power of is , so we must have , which shows

, so also as desired. Therefore

Remark: This is called a dicyclic group. You might consider for which you've seen one of the dicyclic groups before!

Your Response(s):
& Give Up

Problem 5, Part (b) – Give Up – Score: 0 / 7 (39044)

Problem: Report Error

For which values of is the group from the previous part isomorphic to a group
we've met before? You may ignore the possibility that is a nontrivial product of such groups.

1 2 3 4 5

Solution:
For we do get which is a cyclic group of order . For we have
Setting and gives a presentation of the quaternion group.

This clues us in that is not actually a dihedral group in general, so that there's no particular reason we should recognize it for
and in fact we do not. Note that is not Abelian for since we have If we also had
then we'd have so but this does not hold with Then for our only
general candidate for a non-Abelian group of order is If were then the generator would have to represent
a generating rotation. But the second relation shows that could not be a reflection, since every reflection squares to
the identity. Since cannot be generated by rotations, cannot be

There's one last possibility to check: for we know another non-Abelian group of order namely, However,
lacks any element of order so this is ruled out immediately.

These groups are called the dicyclic groups, and are the last nice family of non-Abelian groups we will meet in this course.
Beyond this, one really needs to get into linear algebra over finite fields to continue developing examples.

Your Response(s):
& Give Up

Problem 6 – Completed – Score: 3 / 3 (38749)

Problem: Report Error

Prove that , for any groups and . What is the quotient?

Solution:
Let . Thus, for any and ,

So is a normal subgroup.

The quotient is isomorphic to , via the isomorphism . Since every

coset of has a unique element of the form , this is an isomorphism as desired.

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 4/9
9/10/22, 3:51 AM 3111 Group Theory
Your Response:
Yes

Problem 7 – Give Up – Score: 0 / 7 (38748)

Problem: Report Error

How many elements does have, for ?

Note: Recall that denotes the center of that is, the set of such that for every Be sure to
fully simplify your answer.

Solution:
First: is certainly a normal subgroup, since for any and

Now, has a unique nontrivial central element given by the rotation through radians. This commutes with every rotation,
since all rotations commute, and with every reflection , since for a rotation we have . The same argument
shows that no other rotation commutes with every reflection, and similarly that no reflection commutes with every rotation.

Therefore

Your Response(s):
& Give Up

Problem 8 – Completed – Score: 3 / 3 (38750)

Problem: Report Error

Let and be normal subgroups of of orders and . Prove that contains an element of order

Solution:
Since both subgroups are of prime order, both are cyclic. Let and let . Then by normality of since
there exists some with

Likewise, since is normal, there is some with

Equating these we see

or

Any power of has order or Any power of has order or Therefore

This shows and commute. By the Chinese remainder theorem we have , showing there is an element ( for
example) of order

Your Response:
Yes

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 5/9
9/10/22, 3:51 AM 3111 Group Theory

Problem 9 – Give Up – Score: 0 / 7 (38743)

Problem: Report Error

How many normal subgroups does have?

Solution:
Suppose contains a reflection , and as usual let be a generating rotation in . Then contains all the
conjugates , the four reflections to which is conjugate. Furthermore, contains the rotations
. This gives an eight-element normal subgroup of , the group of symmetries of one of the two squares inscribed
in the octagon. Of course, could also be as a whole.

If contains no reflection, then we claim . All the subgroups of are normal in , but we still have to show that
they're all normal in . Indeed, if then , as we've established in past weeks. Thus
for any and in any subgroup

Thus, has itself and two s as normal subgroups, together with the trivial subgroup, and making for
total.

Your Response(s):
& Give Up

Problem 10 – Give Up – Score: 0 / 7 (38754)

Problem: Report Error

How many distinct quotient groups does have, up to isomorphism?

Solution:
Let . Of course, is the trivial group and is .

Now, 's subgroups and are respectively generated by , and . Its subgroups
are generated by and . Finally, also has a subgroup generated by and .

Since for , these quotients are all s. However, when , we have to check
whether each quotient is or the Klein -group .

We can see that is a , since has order in the quotient group. Indeed, is not in unless
. More geometrically, the cosets of in are the horizontal lines in

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 6/9
9/10/22, 3:51 AM 3111 Group Theory

The quotient group structure on this set of horizontal lines is given by summing the -intercepts, and the group is generated by a
line of -intercept

Similarly, is a generated by and the cosets are lines of slope

However, is a Klein -group! Indeed, and are distinct elements of order in the quotient.
Algebraically, the cosets are sets of the form and we note that there's no obvious algebraic equation
for such a figure. This is reflected geometrically in the fact that each coset is half of some vertical line. Thus there is no single
parameter that determines a coset which might give a generator--one has to specify which vertical line, and which half,
separately, to pick out a coset, and those two specifications correspond to the two generators of the quotient group:

All in all there are isomorphism types of quotients of , namely and , including two
quotients corresponding to a single isomorphism type of subgroup.

Your Response(s):
& Give Up

Problem 11, Part (a) – Give Up – Score: 0 / 7 (38756)

Problem: Report Error

We say is an extension of by if is isomorphic to a normal subgroup such that . Up to
isomorphism, how many Abelian groups can be written as an extension of by ?

Solution:
Such a group must have order We consider the -element Abelian groups , and in turn.

The quotient of by the unique order- subgroup is cyclic of order , as established in class. Meanwhile
is cyclic of order since can only be in if is divisible by , showing that
the coset is of order . However, cannot have as a quotient, since in any group with
and for any we can see that Indeed, if then certainly
Therefore there are such extensions.

Hint(s): What must the order of such a group be?

Your Response(s):
& Give Up

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 7/9
9/10/22, 3:51 AM 3111 Group Theory

Problem 11, Part (b) – Did Not Answer – Technical: 0 / 7 – Style: 0 / 1 (38753)

Problem: Report Error

Which isomorphism classes of Abelian groups are an extension of by ? Recall that we say is an extension
of by if is isomorphic to a normal subgroup and is isomorphic to .

Remember that if you get stuck on a homework problem, you can always ask on the message board! Click on the pencil icon V in
the upper-right corner of the problem, and this will open a box where you can ask your question, which will be posted on the
message board. You can also click on the speech bubble icon t, which will bring up any discussions on that problem.

Solution:
Note that since everything in sight is Abelian, all subgroups are normal, so we can take quotients at will.

First, we see that for any such group, since we have a subgroup of index Thus, by the FTFGAG,
we know where and is cyclic of order If and is nontrivial, then
is divisible by That means that all elements of order are in and in particular,
since contains elements of order Now since is cyclic of order we know must have an element
of order for some and If were nontrivial, then would have order divisible
by so we have and, in particular, has a subgroup which is cyclic of order

Now, since contains , we claim that so that we can ignore the factor of entirely.
Indeed, we have an isomorphism given by sending to It is immediate from the
definition of the group operation on the quotient group that has the homomorphism property. Furthermore, is well-defined.
Indeed, suppose that is, Then for some so that
which shows so that Finally,
is surjective, since is contained in --this means that in
for any and Since and are both of order surjectivity implies injectivity and the
claim has been shown.

We are thus reduced to finding the number of Abelian groups of order containing a cyclic subgroup of order by which the
quotient is isomorphic to We check each possibility for arising from the FTFGAG, namely, could be
or First, any quotient of a cyclic group is cyclic, so we can't have
.

If , then the quotient by is as desired. Indeed, the quotient is an Abelian group of

order in which every element has order at most since for every

Moving on to the case represented as then a cyclic subgroup of order is generated by

some , where without loss of generality has order in (If neither nor had order then would have
order at most ) Then has a multiplicative inverse modulo and we see is also generated by , where
. Thus This implies the coset has order in the
quotient group, since is only in if . Thus the quotient is cyclic, so is not isomorphic to
Note that this is consistent with our argument from class that every quotient of by a subgroup is
another our argument that is a generator is an algebraic analogue of our argument in class that we can generate
the group of lines of a given non-vertical slope in by the line with -intercept

Finally, if then is a quotient: as above, every element of the

quotient is of order or (We could also use the earlier problem on here.) Meanwhile, has no
cyclic subgroups of order since every element has order at most We conclude that the possibilities for are
and .

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 8/9
9/10/22, 3:51 AM 3111 Group Theory
Your Response:
You did not answer this problem.

Copyright © AoPS Incorporated. This page is copyrighted material. You can view and print this page for your own use,
but you cannot share the contents of this page with others.

© 2022 Art of Problem Solving

About Us • Contact Us • Terms • Privacy

Copyright © 2022 Art of Problem Solving

https://artofproblemsolving.com/class/3111-grouptheory/homework/8 9/9