Search for notes by fellow students, in your own course and all over the country.
Browse our notes for titles which look like what you need, you can preview any of the notes via a sample of the contents. After you're happy these are the notes you're after simply pop them into your shopping cart.
Title: Mathematics
Description: This is a 91-page simplified version of algebraic topology. It's concise and self-explanatory and can used to get in-depth understanding without the need of further assistance. The lecture was delivered by H. Wilton in 2015 at Cambridge University. It is organized such that all sub-topics are tabulated on a table of contents for easy navigation. It is expected that students who use this notes must not have problems with algebraic topology going forward.
Description: This is a 91-page simplified version of algebraic topology. It's concise and self-explanatory and can used to get in-depth understanding without the need of further assistance. The lecture was delivered by H. Wilton in 2015 at Cambridge University. It is organized such that all sub-topics are tabulated on a table of contents for easy navigation. It is expected that students who use this notes must not have problems with algebraic topology going forward.
Document Preview
Extracts from the notes are below, to see the PDF you'll receive please use the links above
Part II — Algebraic Topology
Based on lectures by H
...
They are nowhere near accurate representations of what
was actually lectured, and in particular, all errors are almost surely mine
...
The fundamental group of a space, homomorphisms induced by maps of spaces,
change of base point, invariance under homotopy equivalence
...
Path-lifting and homotopy-lifting properties, and
their application to the calculation of fundamental groups
...
*Construction
of the universal covering of a path-connected, locally simply connected space*
...
[5]
The Seifert-Van Kampen theorem
Free groups, generators and relations for groups, free products with amalgamation
...
Applications to the
calculation of fundamental groups
...
[3]
Homology
Simplicial homology, the homology groups of a simplex and its boundary
...
*Proof of functoriality for continuous maps, and of
homotopy invariance*
...
The Mayer-Vietoris theorem
...
Rational homology groups; the
Euler-Poincar´e characteristic and the Lefschetz fixed-point theorem
...
1 Some recollections and conventions
...
2 Cell complexes
...
0 Motivation
...
1 Homotopy
...
2 Paths
...
3 The fundamental group
...
1 Covering space
...
2 The fundamental group of the
3
...
3
...
group
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
its applications
...
...
...
...
...
...
...
...
...
...
...
21
21
29
31
34
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
4 Some group theory
38
4
...
38
4
...
40
4
...
43
5 Seifert-van Kampen theorem
5
...
5
...
5
...
4 The fundamental group of all surfaces
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
1 Simplicial complexes
...
2 Simplicial approximation
...
1 Simplicial homology
...
2 Some homological algebra
...
3 Homology calculations
...
4 Mayer-Vietoris sequence
...
5 Continuous maps and homotopy invariance
...
6 Homology of spheres and applications
...
7 Homology of surfaces
...
8 Rational homology, Euler and Lefschetz numbers
2
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
65
65
70
73
75
82
86
88
91
0
0
Introduction
II Algebraic Topology
Introduction
In topology, a typical problem is that we have two spaces X and Y , and we want
to know if X ∼
= Y , i
...
if X and Y are homeomorphic
...
But what if they
are not? How can we prove that two spaces are not homeomorphic?
For example, are Rm and Rn homeomorphic (for m 6= n)? Intuitively, they
should not be, since they have different dimensions, and in fact they are not
...
It turns out we are much better
at algebra than topology
...
For example, we will be able to reduce the problem of whether Rm
and Rn are homeomorphic (for m =
6 n) to the question of whether Z and {e}
are isomorphic, which is very easy
...
Let Dn be the n dimensional unit disk, and S n−1 be the n−1 dimensional unit
sphere
...
Alternatively, this
says that we cannot continuously map the disk onto the boundary sphere such
that the boundary sphere is fixed by the map
...
This is something we can prove in 5 seconds
...
In algebraic topology, we will be developing a lot of machinery to do this sort
of translation
...
It will take some hard work,
and will be rather tedious and boring at the beginning
...
In case you are completely uninterested in topology, and don’t care if Rm and
n
R are homeomorphic, further applications of algebraic topology include solving
equations
...
If you are not
interested in these either, you may as well drop this course
...
1
II Algebraic Topology
Definitions
Some recollections and conventions
We will start with some preliminary definitions and conventions
...
In this course, the word map will always refer to continuous
maps
...
We are going to build a lot of continuous maps in algebraic topology
...
The gluing lemma tells us that
this works
...
If f : X → Y is a function of topological spaces,
X = C ∪ K, C and K are both closed, then f is continuous if and only if the
restrictions f |C and f |K are continuous
...
Suppose f is continuous
...
So f |C is continuous
...
If f |C and f |K are continuous, then for any closed A ⊆ Y , we have
−1
f −1 (A) = f |−1
C (A) ∪ f |K (A),
which is closed
...
This lemma is also true with “closed” replaced with “open”
...
We will also need the following technical lemma about metric spaces
...
Let (X, d) be a compact metric space
...
Then there is some δ such that for each x ∈ X, there is some α ∈ A
such that Bδ (x) ⊆ Uα
...
Proof
...
Then for each n ∈ N, there is some xn ∈ X such that
B1/n (xn ) is not contained in any Uα
...
Suppose this subsequence converges to y
...
Since Uα
is open, there is some r > 0 such that Br (y) ⊆ Uα
...
But then
B1/n (xn ) ⊆ Br (y) ⊆ Uα
...
1
...
Even if we require them
to be compact, Hausdorff etc, we can often still produce really ugly topological
spaces with weird, unexpected behaviour
...
To build cell complexes, we are not just gluing maps, but spaces
...
For a space X, and a map f : S n−1 → X, the
space obtained by attaching an n-cell to X along f is
X ∪f Dn = (X q Dn )/∼,
where the equivalence relation ∼ is the equivalence relation generated by x ∼ f (x)
for all x ∈ S n−1 ⊆ Dn (and q is the disjoint union)
...
So we are just sticking a disk onto X by attaching the boundary of
the disk onto a sphere within X
...
A (finite) cell complex is a space X obtained by
(i) Start with a discrete finite set X (0)
...
X
= X
q
α
For example, given the X (0) above, we can attach some loops and lines to
obtain the following X (1)
We can add surfaces to obtain the following X (2)
(iii) Stop at some X = X (k)
...
5
1
Definitions
II Algebraic Topology
To define non-finite cell complexes, we just have to remove the words “finite” in
the definition and remove the final stopping condition
...
So instead let’s look at a
non-cell complex
...
The following is not a cell complex: we take R2 , and add a circle with
radius 12 and center (0, 21 )
...
We obtain something like
This is known as the Hawaiian Earring
...
However, in the definition of a cell
complex, the cells are supposed to be completely unrelated and disjoint, apart
from intersecting at the origin
...
In particular, if we take the following sequence (0, 1), (0, 21 ), (0, 14 ), · · · , it
converges to (0, 0)
...
We could have made it such that the nth
cell has radius n
...
We will see that the Hawaiian Earring will be a counterexample to a lot of
our theorems here
...
2
...
Let’s first do the
simple case, where m = 1, n = 2
...
This is not hard
...
Let’s try to
remove a point from each of them
...
However, removing a point does not have this effect on
R2
...
Unfortunately, this does not extend very far
...
What else can we do?
Notice that when we remove a point from R2 , sure it is still connected, but
something has changed
...
If the origin were there,
we can keep shrinking the circle down until it becomes a point
...
The strategy now is to exploit the fact that R2 \ {0} has circles which cannot be
deformed to points
...
1
Homotopy
We have just talked about the notion of “deforming” circles to a point
...
This process of deformation is known as homotopy
...
Notation
...
Definition (Homotopy)
...
A homotopy from f to g
is a map
H :X ×I →Y
such that
H(x, 0) = f (x),
H(x, 1) = g(x)
...
For each time t, H( · , t) defines a map X → Y
...
If such an H exists, we say f is homotopic to g, and write f ' g
...
As mentioned at the beginning, by calling H a map, we are requiring it to
be continuous
...
We might want to make sure
that when we are deforming from a path f to g, the end points of the path don’t
move
...
We say f is homotopic to g rel A, written f '
g rel A, if for all a ∈ A ⊆ X, we have
H(a, t) = f (a) = g(a)
...
Our notation suggests that homotopy is an equivalence relation
...
For spaces X, Y , and A ⊆ X, the “homotopic rel A” relation is
an equivalence relation
...
Proof
...
(ii) Symmetry: if H(x, t) is a homotopy from f to g, then H(x, 1 − t) is a
homotopy from g to f
...
We want to show that f ' h rel A
...
We know how to continuously deform f to g, and from g to h
...
We define H 00 : X × I → Y by
(
H(x, 2t)
0 ≤ t ≤ 12
00
H (x, t) =
H 0 (x, 2t − 1) 12 ≤ t ≤ 1
This is well-defined since H(x, 1) = g(x) = H 0 (x, 0)
...
It is easy to check that H 00 is a homotopy rel A
...
We can extend the notion of homotopy to spaces as well
...
Here, we replace equality by homotopy
...
A map f : X → Y is a homotopy equivalence if there exists a g : Y → X such that f ◦ g ' idY and g ◦ f ' idX
...
If a homotopy equivalence f : X → Y exists, we say that X and Y are
homotopy equivalent and write X ' Y
...
Clearly,
homeomorphic spaces are homotopy equivalent
...
Example
...
We have a natural inclusion map
i : X ,→ Y
...
y
r(y)
In particular, we define r : Y → X by
r(y) =
y
...
We will now see that i ◦ r ' idY
...
So this is just the projection map
...
t + (1 − t)kyk
This is continuous, and H( · , 0) = i ◦ r, H( · , 1) = idY
...
We started with a 2-dimensional R2 \ {0} space, and squashed it into a
one-dimensional sphere
...
Dimensions seem to be a rather fundamental thing in geometry, and we are
discarding it here
...
We will later see that this is what homotopy equivalence
preserves
...
Let Y = Rn , X = {0} = ∗
...
Again, we have
r ◦ i = idX
...
Again, from the point of view of homotopy theory, Rn is just the same as a
point! You might think that this is something crazy to do — we have just given
up a lot of structure of topological spaces
...
For example,
9
2
Homotopy and the fundamental group
II Algebraic Topology
it is often much easier to argue about the one-point space ∗ than the whole of
R2 ! By studying properties that are preserved by homotopy equivalence, and
not just homeomorphism, we can simplify our problems by reducing complicated
spaces to simpler ones via homotopy equivalence
...
Notation
...
Definition (Contractible space)
...
We now show that homotopy equivalence of spaces is an equivalence relation
...
Lemma
...
Proof
...
Then we are done since
homotopy between maps is an equivalence relation
...
(i) Consider the following composition:
X ×I
H
Y
g0
Z
It is easy to check that this is the first homotopy we need to show g0 ◦ f0 '
g0 ◦ f1
...
Homotopy equivalence of spaces is an equivalence relation
...
Symmetry and reflexivity are trivial
...
We will show that h ◦ f : X → Z is a homotopy
equivalence with homotopy inverse g ◦ k
...
Similarly,
(g ◦ k) ◦ (h ◦ f ) = g ◦ (k ◦ h) ◦ f ' g ◦ idY ◦ f = g ◦ f ' idX
...
Definition (Retraction)
...
A retraction r : X → A
is a map such that r ◦ i = idA , where i : A ,→ X is the inclusion
...
10
2
Homotopy and the fundamental group
II Algebraic Topology
This map sends everything in X to A without moving things in A
...
Definition (Deformation retraction)
...
A deformation retraction is strong if we require this homotopy
to be a homotopy rel A
...
Example
...
Then the constant map
r : X → A is a retraction
...
2
...
A path in a space X is a map γ : I → X
...
If γ(0) = γ(1), then γ is called a loop (based at x0 )
...
It can be self-intersecting
or do all sorts of weird stuff
...
In homotopy
theory, we do so using the idea of paths
...
Definition (Concatenation of paths)
...
This is continuous by the gluing lemma
...
x0
x2
x1
11
2
Homotopy and the fundamental group
II Algebraic Topology
Definition (Inverse of path)
...
This is exactly the same path but going in the opposite direction
...
Definition (Constant path)
...
We haven’t actually got a good algebraic system
...
Also, we are not able to combine arbitrary paths in a space
...
We can view this as a first attempt at associating
things to topological spaces
...
We can define a relation on X: x1 ∼ x2 if there
exists a path from x1 to x2
...
The equivalence classes [x] are called path components
...
In the above space, we have three path components
...
e
...
However, this is a first step
at associating something to spaces
...
One important property of this π0 is that not only does it associate a set to
each topological space, but also associates a function between the corresponding
sets to each continuous map
...
For any map f : X → Y , there is a well-defined function
π0 (f ) : π0 (X) → π0 (Y ),
defined by
π0 (f )([x]) = [f (x)]
...
(ii) For any maps A
h
B
k
C , we have π0 (k ◦ h) = π0 (k) ◦ π0 (h)
...
To show this is well-defined, suppose [x] = [y]
...
Then f ◦ γ is a path from f (x) to f (y)
...
(i) If f ' g, let H : X × I → Y be a homotopy from f to g
...
Then
H(x, · ) is a path from f (x) to g(x)
...
e
...
So π0 (f ) = π0 (g)
...
(iii) π0 (idX )([x]) = [idX (x)] = [x]
...
Corollary
...
Example
...
This is a rather silly example, since we can easily prove it directly
...
Now let’s return to our operations on paths, and try to make them algebraic
...
Paths γ, γ 0 : I → X are homotopic as paths if
they are homotopic rel {0, 1} ⊆ I, i
...
the end points are fixed
...
x0
x1
Note that we would necessarily want to fix the two end points
...
This homotopy works well with our previous operations on paths
...
Let γ1 , γ2 : I → X be paths, γ1 (1) = γ2 (0)
...
γ10
γ20
x0
x2
x1
γ1
γ2
Proof
...
Then we have the diagram
γ10
x0
H1
γ20
x1
γ1
H2
γ2
13
x2
2
Homotopy and the fundamental group
II Algebraic Topology
We can thus construct a homotopy by
(
H1 (s, 2t)
0 ≤ t ≤ 12
H(s, t) =
...
Proposition
...
Then
(i) (γ0 · γ1 ) · γ2 ' γ0 · (γ1 · γ2 )
(ii) γ0 · cx1 ' γ0 ' cx0 · γ0
...
Proof
...
14
2
Homotopy and the fundamental group
2
...
We want to try to do
this using paths
...
According to our proposition,
this operation satisfies associativity, inverses and identity
...
The idea is to fix one of the points x0 in our space, and only think about
loops that start and end at x0
...
This tells us that we aren’t going to just think about spaces, but spaces with
basepoints
...
Definition (Fundamental group)
...
The fundamental group of X (based at x0 ), denoted π1 (X, x0 ), is the set of homotopy
classes of loops in X based at x0 (i
...
γ(0) = γ(1) = x0 )
...
Often, when we write the homotopy classes of paths [γ], we just get lazy and
write γ
...
The fundamental group is a group
...
Immediate from our previous lemmas
...
Unfortunately, it is rather difficult to prove that a space has a non-trivial
fundamental group, until we have developed some relevant machinery
...
Instead,
we will look at some properties of the fundamental group first
...
A based space is a pair (X, x0 ) of a space X and a
point x0 ∈ X, the basepoint
...
A based homotopy is a
homotopy rel {x0 }
...
We can do the same for π1
...
To a based map
f : (X, x0 ) → (Y, y0 ),
there is an associated function
f∗ = π1 (f ) : π1 (X, x0 ) → π1 (Y, y0 ),
defined by [γ] 7→ [f ◦ γ]
...
15
2
Homotopy and the fundamental group
II Algebraic Topology
(ii) If f ' f 0 , then π1 (f ) = π1 (f 0 )
...
(B, b)
k
(C, c) , we have π1 (k ◦ h) =
(iv) π1 (idX ) = idπ1 (X,x0 )
Proof
...
In category-theoretic language, we say that π1 is a functor
...
However, to define the fundamental group, we had to make a
compromise and pick a basepoint
...
We don’t
want a basepoint! Hence, we should look carefully at what happens when we
change the basepoint, and see if we can live without it
...
Hence,
the first importance of picking a basepoint is picking a path component
...
Now we want to compare fundamental groups with different basepoints
...
Suppose we have a loop γ at x0
...
We first pick a path
u : x0
x1
...
e
...
x0
x1
u
Proposition
...
This satisfies
(i) If u ' u0 , then u# = u0#
...
Then (u · v)# = v# ◦ u#
...
16
2
Homotopy and the fundamental group
II Algebraic Topology
A nicer way of writing this is
f∗
π1 (X, x0 )
u#
π1 (Y, y0 )
(f ◦u)#
f∗
π1 (X, x1 )
π1 (Y, y1 )
The property says that the composition is the same no matter which way
we go from π1 (X, x0 ) to π1 (Y, y1 )
...
These diagrams will appear all of the time in this course
...
It is important (yet difficult) to get the order of concatenation and composition
right
...
Proof
...
Note that (u−1 )# = (u# )−1 , which is why we have
an isomorphism
...
So the basepoint isn’t really too important
...
While the two groups are isomorphic, the actual isomorphism depends on which
path u : x0
x1 we pick
...
In particular, we cannot say “let α ∈ π1 (X, x0 )
...
We can also see that if (X, x0 ) and (Y, y0 ) are based homotopy equivalent,
then π1 (X, x0 ) ∼
= π1 (Y, y0 )
...
So
f∗ ◦ g∗ = idπ1 (Y,y0 ) ,
g∗ ◦ f∗ = idπ1 (X,x0 ) ,
and f∗ and g∗ are the isomorphisms we need
...
If this were a based homotopy, we know
f∗ = g∗ : π1 (X, x0 ) → π1 (Y, f (x0 ))
...
How can we relate f∗ : π1 (X, x0 ) → π1 (Y, f (x0 )) and g∗ : π1 (X, x0 ) →
π1 (Y, g(x0 ))?
First of all, we need to relate the groups π1 (Y, f (x0 )) and π1 (Y, g(x0 ))
...
To produce this, we can use
the homotopy H
...
Now we have
three maps f∗ , g∗ and u#
...
17
2
Homotopy and the fundamental group
X
II Algebraic Topology
Y
f
f (x0 )
x0
g(x0 )
g
Lemma
...
Proof
...
We need to check that
g∗ ([γ]) = u# ◦ f∗ ([γ])
...
To prove this result, we want to build a homotopy
...
Our plan is to exhibit two homotopic paths `+ and `− in I × I such that
F ◦ `+ = g ◦ γ,
F ◦ `− = u−1 · (f ◦ γ) · u
...
So to construct a homotopy, we make ourselves work in a
much nicer space I × I
...
18
2
Homotopy and the fundamental group
II Algebraic Topology
I ×I
`+
Y
g
F
`−
u−1
u
f
More precisely, `+ is the path s 7→ (s, 1), and `− is the concatenation of the
paths s 7→ (0, 1 − s), s 7→ (s, 0) and s 7→ (1, s)
...
If this is not obvious, we can
manually check the homotopy
L(s, t) = t`+ (s) + (1 − t)`− (s)
...
Hence F ◦ `+ 'F ◦L F ◦ `− as paths
...
We
have
F ◦ `+ (s) = H(γ(s), 1) = g ◦ γ(s)
...
So done
...
With this lemma, we can show that fundamental groups really respect
homotopies
...
If f : X → Y is a homotopy equivalence, and x0 ∈ X, then the
induced map
f∗ : π1 (X, x0 ) → π1 (Y, f (x0 ))
...
While this seems rather obvious, it is actually non-trivial if we want to do
it from scratch
...
So this proof involves
some real work
...
Let g : Y → X be a homotopy inverse
...
19
2
Homotopy and the fundamental group
X
II Algebraic Topology
Y
f
x0
f (x0 )
0
u
g ◦ f (x0 )
g
We have no guarantee that g ◦ f (x0 ) = x0 , but we know that our homotopy H 0
gives us u0 = H 0 (x0 , · ) : x0
g ◦ f (x0 )
...
However, we know that u0# is an isomorphism
...
Doing it the other way round with f ◦ g instead of g ◦ f , we know that g∗ is
injective and f∗ is surjective
...
With this theorem, we can finally be sure that the fundamental group
is a property of the space, without regards to the basepoint (assuming path
connectedness), and is preserved by arbitrary homotopies
...
Definition (Simply connected space)
...
Example
...
Hence, any contractible space is simply connected since
it is homotopic to ∗
...
There is a useful characterization of simply connected spaces:
Lemma
...
Proof
...
Now
note that u · v −1 is a loop based at x0 , it is homotopic to the constant path, and
v −1 · v is trivially homotopic to the constant path
...
On the other hand, suppose there is a unique homotopy class of paths x0
x1
for all x0 , x1 ∈ X
...
So π1 (X, x0 ) is trivial
...
Groups were created to represent symmetries of objects
...
If not, something wrong is probably going on
...
So what do they act on? Can we
find something on which these fundamental groups act on?
An answer to this would also be helpful in more practical terms
...
This would be easy if we
can make the group act on something — if the group acts non-trivially on our
thing, then clearly the group cannot be trivial
...
3
...
˜
X
p
U
x0
X
˜ p:X
˜ → X),
Definition (Covering space)
...
Whether we require p to be surjective is a matter of taste
...
˜ → X if
Definition (Evenly covered)
...
where p|Vα : Vα → U is a homeomorphism, and each of the Vα ⊆ X
Example
...
Duh
...
Consider p : R → S 1 ⊆ C defined by t 7→ e2πit
...
21
3
Covering spaces
II Algebraic Topology
R
p−1 (1)
p
1
S1
Here we have p−1 (1) = Z
...
Before we do that, we can guess what the fundamental
group of S 1 is
...
So we can characterize each loop by
the number of times it loops around the circle, and it should not be difficult to
convince ourselves that two loops that loop around the same number of times
can be continuously deformed to one another
...
However, we also know that p−1 (1) = Z
...
We will show that this is not
a coincidence
...
Consider pn : S 1 → S 1 (for any n ∈ Z \ {0}) defined by z 7→ z n
...
This time the pre-image of 1 would be n copies of 1, instead of Z copies of 1
...
Consider X = RP2 , which is the real projective plane
...
e
...
We can also think of this as the space of lines in R3
...
There is, however, a more convenient way of thinking about RP2
...
22
3
Covering spaces
II Algebraic Topology
U
U
We define p : S 2 → RP2 to be the quotient map
...
As we mentioned, a covering space of X is (locally) like many “copies” of
X
...
˜
“lift” it to a function f : Y → X,
This is not always possible, but when it is, we call this a lift
...
Let f : Y → X be a map, and p : X
˜ such that f = p ◦ f˜, i
...
the following
space
...
f˜(Y )
f˜
˜
X
p
f (Y )
Y
X
f
It feels that if we know which “copy” of X we lifted our map to, then we
already know everything about f˜, since we are just moving our f from X to
that particular copy of X
...
Let p : X
be both lifts of f
...
In particular, if Y is connected, f˜1 and f˜2 agree either
everywhere or nowhere
...
If we know a point in the
lift, then we know the whole path
...
e
...
Proof
...
Let y be such that f˜1 (y) = f˜2 (y)
...
Let U
˜ , p(U
˜ ) = U and p| ˜ : U
˜ → U is a homeomorphism
...
We will show that f˜1 = f˜2 on V
...
Since p|U˜ is a homeomorphism, it follows that
f˜1 |V = f˜2 |V
...
Suppose not
...
So f˜1`
(y) 6=
f˜2 (y)
...
Let p−1 (U ) = Uα
...
Then V = f˜1−1 (Uβ ) ∩ f˜2−1 (Uγ ) is
an open neighbourhood of y, and hence intersects S by definition of closure
...
But f˜1 (x) ∈ Uβ and f˜2 (x) ∈ Uγ ,
and hence Uβ and Uγ have a non-trivial intersection
...
So
S is closed
...
How about existence? Given a map,
is there guarantee that we can lift it to something? Moreover, if I have fixed a
“copy” of X I like, can I also lift my map to that copy? We will later come up
with a general criterion for when lifts exist
...
˜ → X be a covering space,
Lemma (Homotopy lifting lemma)
...
Let f˜0 be a lift of f0
...
e
...
(ii) H
This lemma might be difficult to comprehend at first
...
Then a homotopy is just a path
...
Let p : X
˜ such that p(˜
path, and x
˜0 ∈ X
x0 ) = x0 = γ(0)
...
e
...
This is exactly the picture we were drawing before
...
Note that we have already proved uniqueness
...
24
3
Covering spaces
II Algebraic Topology
In theory, it makes sense to prove homotopy lifting, and path lifting comes
immediately as a corollary
...
So instead, we will prove path lifting, which is something we can more
easily visualize and understand, and then use that to prove homotopy lifting
...
Let
S = {s ∈ I : γ˜ exists on [0, s] ⊆ I}
...
(ii) S is open
...
If s 6∈ S, then pick an evenly covered neighbourhood U of γ(s)
...
So s − 2ε 6∈ S
...
So S is
closed
...
So γ˜
exists
...
So H(y,
·)
is defined
...
Steps
of the proof are
(i) Use compactness of I to argue that the proof of path lifting works on small
neighbourhoods in Y
...
(iii) By uniqueness of lifts, these path liftings agree when they overlap
...
With the homotopy lifting lemma in our toolkit, we can start to use it to
do stuff
...
We are now
going to build a bridge between these two, and show how covering spaces can be
used to reflect some structures of the fundamental group
...
We have just showed that we are allowed to lift homotopies
...
The homotopy lifting lemma
does not tell us that the lifted homotopy preserves basepoints
...
˜ are lifts
Corollary
...
If γ ' γ 0 as paths, then γ˜ and γ˜ 0 are homotopic as paths
...
25
3
Covering spaces
II Algebraic Topology
Note that if we cover the words “as paths” and just talk about homotopies,
then this is just the homotopy lifting lemma
...
˜ a lift of H with H(
˜ · , 0) = γ˜
...
The homotopy lifting lemma gives us an H,
γ˜0
γ0
cx 0
H
lift
cx 1
cx˜0
γ
˜
H
cx˜1
γ˜
˜ square is γ˜
...
˜ · , 1) is a lift of H( · , 1) = γ 0 , starting at x
Now H(
˜0
...
So this is indeed a homotopy between γ˜ and γ˜ 0
...
˜
We know that H(0,
· ) is a lift of H(0, · ) = cx0
...
By uniqueness of lifts, we must have H(0,
· ) = cx˜0
...
So this is a homotopy of paths
...
is it? Is it possible that we have four copies of x0 but just three copies
of x1 ? This is obviously possible if X is not path connected — the component
containing x0 and the one containing x1 are completely unrelated
...
If X is a path connected space, x0 , x1 ∈ X, then there is a bijection
p−1 (x0 ) → p−1 (x1 )
...
We want to use this to construct a bijection
Proof
...
The obvious thing to do
is to use lifts of the path γ
...
The inverse map is obtained by replacing γ with γ −1 , i
...
fγ −1
...
Now
notice that γ˜ −1 is a lift of γ −1 starting at x
˜1 and ending at x
˜0
...
So fγ −1 is an inverse to fγ , and hence fγ is bijective
...
A covering space p : X
−1
X is n-sheeted if |p (x)| = n for any (and hence all) x ∈ X
...
Is there any number we can assign to fundamental groups? Well, the
index of a subgroup might be a good candidate
...
One important property of covering spaces is the following:
˜ → X is a covering map and x
˜ then
Lemma
...
Proof
...
˜ We let γ = p ◦ γ˜
...
e
...
0
paths, the homotopy lifting lemma then gives us a homotopy upstairs between γ˜
and cx˜0
...
As we have originally said, our objective is to make our fundamental group
act on something
...
Let’s look again at the proof that there is a bijection between p−1 (x0 ) and
−1
p (x1 )
...
This end point may or may not be our original x
˜0
...
However, we are not really interested in paths themselves
...
However, this is fine
...
In particular, these have the same end points
...
27
3
Covering spaces
II Algebraic Topology
˜
X
x
˜0
γ˜
x
˜00
p
γ
x0
X
Now this gives an action of π1 (X, x0 ) on p−1 (x0 )! Note, however, that this will
not be the sort of actions we are familiar with
...
To see this, we have to consider what happens when we
perform two operations one after another, which you shall check yourself
...
When we have an action, we are interested in two things — the orbits, and
the stabilizers
...
Lemma
...
˜ is path
(i) The action of π1 (X, x0 ) on p−1 (x0 ) is transitive if and only if X
connected
...
˜ x
(ii) The stabilizer of x
˜0 ∈ p−1 (x0 ) is p∗ (π1 (X,
˜0 )) ⊆ π1 (X, x0 )
...
˜ x
Note that p∗ π1 (X,
˜0 )\π1 (X, x0 ) is not a quotient, but simply the set of
cosets
...
Note that this is great! If we can find a covering space p and a point x0
such that p−1 (x0 ) is non-trivial, then we immediately know that π1 (X, x0 ) is
non-trivial!
Proof
...
Then we can project this to γ = p ◦ γ˜
...
e
...
Then by the definition of the action,
x
˜0 · [γ] = γ˜ (1) = x
˜00
...
Then γ˜ is a loop based at x˜0
...
(iii) This follows directly from the orbit-stabilizer theorem
...
We can be more ambitious, and try to actually find π1 (X, x0 )
...
Then we have a
bijection between π1 (X, x0 ) and p−1 (x0 )
...
space X
˜ → X is a universal cover
Definition (Universal cover)
...
if X
We will look into universal covers in depth later and see what they really are
...
If p : X
π1 (X, x0 ) → p−1 (x0 )
...
To obtain a bijection, we need to
pick a starting point x
˜0 ∈ p−1 (x0 )
...
3
...
We are going to consider
the space S 1 and a universal covering R
...
There is a bijection π1 (S 1 , 1) → p−1 (1) = Z
...
The map ` : π1 (S 1 , 1) → p−1 (1) = Z is a group isomorphism
...
We know it is a bijection
...
The idea is to write down representatives for what we think the elements should
be
...
Since R is simply
connected, there is a unique homotopy class between any two points
...
So
[γ] = [un ]
...
Therefore
m · un = u
`([um ][un ]) = `([um · um ]) = m + n = `([um+n ])
...
What have we done? In general, we might be given a horrible, crazy loop
in S 1
...
So we pull it
up to the universal covering R
...
We then project
this homotopy down to S 1 , to get a homotopy from γ to un
...
With the fundamental group of the circle, we do many things
...
Since C \ {0} is homotopy equivalent to S 1 , its fundamental group is Z
as well
...
Any such group homomorphism must be of the form t 7→ nt, and the
winding number is given by n
...
Also, we have the following classic application:
Theorem (Brouwer’s fixed point theorem)
...
If f : D2 → D2 is continuous, then there is some x ∈ D2 such
that f (x) = x
...
Suppose not
...
30
3
Covering spaces
II Algebraic Topology
f (x)
x
g(x)
We define g : D2 → S 1 as in the picture above
...
In other words, the following
composition is the identity:
S1
ι
D2
g
S1
idS 1
Then this induces a homomorphism of groups whose composition is the identity:
Z
ι∗
{0}
g∗
Z
idZ
But this is clearly nonsense! So we must have had a fixed point
...
What about D3 ? Can we prove a similar theorem?
Here the fundamental group is of little use, since we can show that the fundamental group of S n for n ≥ 2 is trivial
...
3
...
We have just shown that p : R → S 1 is a universal cover
...
What would be a
universal cover of the torus S 1 × S 1 ? An obvious guess would be p × p : R × R →
S 1 × S 1
...
Alternatively, we can see it as a quotient of the square,
where we identify the following edges:
31
3
Covering spaces
II Algebraic Topology
Then what does it feel like to live in the torus? If you live in a torus and look
around, you don’t see a boundary
...
The difference is that in the torus, you aren’t actually seeing
free space out there, but just seeing copies of the same space over and over again
...
Whenever we move one unit
horizontally or vertically, we get back to “the same place”
...
This space has a huge translation symmetry
...
We see that if we live inside the torus S 1 × S 1 , it feels like we are actually
living in the universal covering space R × R, except that we have an additional
symmetry given by the fundamental group Z × Z
...
We would like to
say that universal covers always exist
...
Firstly, we should think — what would having a universal cover imply?
˜ Pick any point x0 ∈ X, and pick an
Suppose X has a universal cover X
...
˜ If we draw a
evenly covered neighbourhood U in X
...
But we know
˜ is simply connected
...
Hence γ
that X
is also homotopic to the constant path
...
It seems like for every x0 ∈ X, there is some neighbourhood of x0 that is
simply connected
...
The homotopy
˜ and can pass through anything
from γ˜ to the constant path is a homotopy in X,
32
3
Covering spaces
II Algebraic Topology
˜ not just U
˜
...
So U itself need not be simply connected
...
Definition (Locally simply connected)
...
As we mentioned, what we actually want is a weaker condition
...
X is semi-locally simply connected
if for all x0 ∈ X, there is some neighbourhood U of x0 such that any loop γ
based at x0 is homotopic to cx0 as paths in X
...
This is really not interesting, since we don’t care
if a space is semi-locally simply connected
...
We still need one more additional condition:
Definition (Locally path connected)
...
It is important to note that a path connected space need not be locally path
connected
...
Theorem
...
Note that we can alternatively define a universal covering as a covering space
of X that is also a covering space of all other covers of X
...
However, that proof
is not too helpful since it does not tell us where the universal covering comes
from
...
Proof
...
Suppose we have a
˜ Then this lifts to some x
˜ If we have any other point
universal covering X
...
˜ since X
˜ should be path connected, there is a path α
x
˜ ∈ X,
˜:x
˜0
x
˜
...
have another path, then since X
˜ with a path from x
Hence, we can identify each point in X
˜0 , i
...
˜ ←→ {paths α
˜
{points of X}
˜ from x
˜0 ∈ X}/'
...
˜
This is not too helpful though, since we are defining X
˜
However, by path lifting, we know that paths α
˜ from x
˜0 in X biject with paths
α from x0 in X
...
˜ ←→ {paths α from x0 ∈ X}/'
...
X
˜ → X is given by [α] 7→ α(1)
...
33
3
Covering spaces
3
...
We have already established the result
that covering maps are injective on π1
...
It
turns out that as long as we define carefully what we mean for based covering
spaces to be “the same”, this is a one-to-one correspondence — each subgroup
corresponds to a covering space
...
Recall that we have shown that π1 (X, x0 ) acts on p−1 (x0 )
...
Having groups acting on a cube is fun because the cube has some structure
...
We note that we can make π1 (X, x0 ) “act on” the universal cover
...
In general, a point
˜ can be thought of as a path α on X starting from x0
...
α
˜
X
x
˜00
x
˜0 γ˜
p
γ
α
x0
X
We will use this idea and return to the initial issue of making subgroups correspond to covering spaces
...
We want to say “For any subgroup H ≤ π1 (X, x0 ),
˜ x
˜ x
there is a based covering map p : (X,
˜0 ) → (X, x0 ) such that p∗ π1 (X,
˜0 ) = H”
...
So we need some
additional assumptions
...
Let X be a path connected, locally path connected and semilocally simply connected space
...
34
3
Covering spaces
II Algebraic Topology
Proof
...
We have an intermediate
simply connected space, let X
˜ x
group H such that π1 (X,
˜0 ) = 1 ≤ H ≤ π1 (X, x0 )
...
Now we
action of π1 (X, x0 )
...
We define ∼H on X
the orbit relation for the action of H, i
...
x
˜ ∼H y˜ if there is some h ∈ H such
˜ be the quotient space X/∼
¯ H
...
We then let X
We can now do the messy algebra to show that this is the covering space we
want
...
e
...
Now we want to prove injectivity
...
˜ x
Suppose we have path-connected spaces (Y, y0 ), (X, x0 ) and (X,
˜0 ), with
˜ x
˜0 ) → (X, x0 ) a covering map
...
˜ x
˜0 ) → (X, x0 ) be a covering map of pathLemma (Lifting criterion)
...
If f : (Y, y0 ) → (X, x0 ) is a continuous map, then there is a (unique)
˜ x
lift f˜ : (Y, y0 ) → (X,
˜0 ) such that the diagram below commutes (i
...
p ◦ f˜ = f ):
˜ x
(X,
˜0 )
f˜
(Y, y0 )
f
p
(X, x0 )
if and only if the following condition holds:
˜ x
f∗ π1 (Y, y0 ) ≤ p∗ π1 (X,
˜0 )
...
So this lemma is
really about existence
...
g
...
So paths and
homotopies can always be lifted
...
One direction is easy: if f˜ exists, then f = p ◦ f˜
...
So we
know that im f∗ ⊆ im p∗
...
In the other direction, uniqueness follows from the uniqueness of lifts
...
We define f˜ as follows:
Given a y ∈ Y , there is some path αy : y0
y
...
By path lifting, this path lifts uniquely to β˜y in X
...
Note that if f exists, then this must be what f sends y to
...
35
3
Covering spaces
II Algebraic Topology
Suppose we picked a different path αy0 : y0
y
...
˜ x
˜0 ) says f ◦ γ is the image of a
Our condition that f∗ π1 (Y, y0 ) ≤ p∗ π1 (X,
0
˜
˜
˜
˜ and hence have the same
loop in X
...
So this shows that f˜ is well-defined
...
First, observe that any open set U ⊆ X
can be written as a union of V˜ such that p|V˜ : V˜ → p(V˜ ) is a homeomorphism
...
Let y ∈ f˜−1 (V˜ ), and let x = f (y)
...
We claim that W ⊆ f˜−1 (V˜ )
...
Then f sends
˜ is given by p|−1 (f (γ)),
this to a path from x to f (z)
...
So it follows that f˜(z) = p|V−1
(f
(z))
∈ V˜
...
What this statement properly means is made precise in the following
proposition:
˜1, x
˜2, X
˜ 2 ) be path-connected based spaced,
˜1 ), (X
Proposition
...
Then we have
˜1, x
˜ 2 , x˜2 )
p1∗ π1 (X
˜1 ) = p2∗ π1 (X
if and only if there is some homeomorphism h such that the following diagram
commutes:
h
˜1, x
˜2, x
(X
˜1 )
(X
˜2 )
p1
p2
(X, x0 )
i
...
p1 = p2 ◦ h
...
We are saying that we can find a nice homeomorphism
that works well with the covering map p
...
If such a homeomorphism exists, then clearly the subgroups are equal
...
By symmetry, we can get h−1 = p˜2
...
36
3
Covering spaces
II Algebraic Topology
Since idX˜ 1 is also a lift, by the uniqueness of lifts, we know p˜2 ◦ p˜1 is the identity
map
...
˜1, x
(X
˜1 )
p˜2
˜1, x
(X
˜1 )
p˜1
˜2, x
(X
˜2 )
p2
p1
(X, x0 )
p1
Now what we would like to do is to forget about the basepoints
...
This doesn’t actually change the
group itself, but if we are talking about subgroups, conjugation can send a
subgroup into a different subgroup
...
Proposition
...
37
4
Some group theory
4
II Algebraic Topology
Some group theory
Algebraic topology is about translating topology into group theory
...
Well, maybe you do, but not the right
group theory
...
1
Free groups and presentations
Recall that in IA Groups, we defined, say, the dihedral group to be
D2n = hr, s | rn = s2 = e, srs = r−1 i
...
This is, in some sense, the
“freest” group we can have
...
Definition (Alphabet and words)
...
We assume that
S ∩ S −1 = ∅
...
e
...
Example
...
Then words could be the empty word ∅, or a, or
aba−1 b−1 , or aa−1 aaaaabbbb, etc
...
When we see things like aa−1 , we would want to cancel them
...
Definition (Elementary reduction)
...
Since each reduction shortens the word, and the word is finite in length, we
cannot keep reducing for ever
...
Definition (Reduced word)
...
Example
...
Note that there is an inclusion map S → S ∗ that sends the symbol sα to the
word sα
...
The free group on the set S, written F (S), is the set
of reduced words on S ∗ together with some operations:
(i) Multiplication is given by concatenation followed by elementary reduction
to get a reduced word
...
−1
−1 −1
(iii) The inverse of x1 · · · xn is x−1
= sα
...
Note that we have not showed that multiplication is well-defined — we might
reduce the same word in different ways and reach two different reduced words
...
This is a cleaner
way to define the free group without messing with alphabets and words, but is
(for most people) less intuitive
...
We will state this definition as a lemma
...
If G is a group and φ : S → G is a set map, then there exists a unique
homomorphism f : F (S) → G such that the following diagram commutes:
F (S)
f
S
φ
G
where the arrow not labeled is the natural inclusion map that sends sα (as a
symbol from the alphabet) to sα (as a word)
...
Clearly if f exists, then f must send each sα to φ(sα ) and s−1
...
So if f exists, it must be unique
...
This is well-defined if we define F (S) to be the set of all reduced words, since
each reduced word has a unique representation (since it is defined to be the
representation itself)
...
Then
xy = x1 · · · xn y1 · · · ym
...
So f is a homomorphism
...
We can show that F (S) is the
unique group satisfying the conditions of this lemma (up to isomorphism), by
taking G = F (S) and using the uniqueness properties
...
Let S be a set, and let R ⊆ F (S) be any
subset
...
e
...
This can be given explicitly by
( n
)
Y
−1
hhRii =
gi ri gi : n ∈ N, ri ∈ R, gi ∈ F (S)
...
This is just the usual notation we have for groups
...
Again, we can define this with a universal property
...
If G is a group and φ : S → G is a set map such that f (r) = 1 for all
±1
±1
±1
r ∈ R (i
...
if r = s±1
φ(s2 )±1 · · · φ(sm )±1 = 1),
1 s2 · · · sm , then φ(r) = φ(s1 )
then there exists a unique homomorphism f : hS | Ri → G such that the
following triangle commutes:
hS | Ri
f
S
φ
G
Proof is similar to the previous one
...
Example (The stupid canonical presentation)
...
We can view
the group as a set, and hence obtain a free group F (G)
...
Then by the universal property of the free group, there is a
surjection f : F (G) → G
...
Then hG | Ri is a presentation
for G, since the first isomorphism theorem says
G∼
= F (G)/ ker f
...
For example, even the simplest non-trivial
group Z/2 will be written as a quotient of a free group with two generators
...
Example
...
Example
...
4
...
We are now going to do this using topology
...
For the following illustration, we will just assume
S = {a, b}, but what we will do works for any set S
...
This is a cell complex, with one 0-cell and |S|
1-cells
...
We will call the 0-cells and 1-cells vertices and
edges, and call the whole thing a graph
...
e
...
Moreover, this graph is connected
and simply connected, i
...
it’s a tree
...
So it must have 4
edges attached to it
...
So X
˜ looks like this:
out
...
It is easy to show this is really a covering map
...
Notice that
˜ starting at x
every word w ∈ S ∗ denotes a unique “edge path” in X
˜0 , where an
edge path is a sequence of oriented edges e˜1 , · · · , e˜n such that the “origin” of
e˜i+1 is equal to the “terminus” of e˜i
...
x
˜0
We can note a few things:
˜ is connected
...
(ii) If an edge-path γ˜ fails to be locally injective, we can simplify it
...
So it fails to be locally injective if two
consecutive edges in the path are the same edge with opposite orientations:
e˜i
e˜i+1
e˜i−1
e˜i+2
We can just remove the redundant lines and get
e˜i−1
e˜i+2
This reminds us of two things — homotopy of paths and elementary
reduction of words
...
(iv) For any w ∈ S ∗ , from (ii), we know that γ˜ is locally injective if and only if
w is reduced
...
So there is a bijection between F (S) and π1 (X, x0 )
...
So this bijection identifies the two group structures
...
4
...
But we want
to find the fundamental group of more things
...
Cell complexes are formed by gluing things together
...
Suppose a space X is constructed from two spaces A, B by gluing (i
...
X = A ∪ B)
...
To
understand this, we need to understand how to “glue” groups together
...
Suppose we have groups G1 = hS1 | R1 i, G2 = hS2 |
R2 i, where we assume S1 ∩ S2 = ∅
...
This is not a really satisfactory definition
...
However, it is clear that
this group exists
...
Then we can have the following universal
property of the free product:
Lemma
...
It is immediate from the universal property of the definition of presentations
...
The free product is well-defined
...
The conclusion of the universal property can be seen to characterize
G1 ∗ G2 up to isomorphism
...
Combining the two would give everything we want
...
In terms of gluing spaces, this corresponds to
gluing A and B when A ∩ B is trivial (i
...
simply connected)
...
Definition (Free product with amalgamation)
...
H
Here we are attempting to identify things “in H” as the same, but we need
to use the maps jk and ik to map the things from H to G1 ∗ G2
...
So we
quotient by the (normal closure) of these things
...
G1 ∗ G2 is the group such that for any group K and homomorphisms
H
φi : Gi → K, there exists a unique homomorphism G1 ∗ G2 → K such that the
H
following diagram commutes:
H
i2
G2
i1
G1
j2
j1
φ2
G1 ∗ G2
H
f
φ1
K
This is the language we will need to compute fundamental groups
...
44
5
Seifert-van Kampen theorem
5
5
...
Here we let X = A ∪ B, where A, B, A ∩ B are path-connected
...
Since we like diagrams, we can
write this as a commutative diagram:
A∩B
B
A
X
where all arrows are inclusion (i
...
injective) maps
...
Then we have the
induced homomorphisms
π1 (A ∩ B, x0 )
π1 (B, x0 )
π1 (A, x0 )
π1 (X, x0 )
We might guess that π1 (X, x0 ) is just the free product with amalgamation
π1 (X, x0 ) = π1 (A, x0 )
∗
π1 (A∩B,x0 )
π1 (B, x0 )
...
Theorem (Seifert-van Kampen theorem)
...
Then for any x0 ∈ A ∩ B,
we have
π1 (X, x0 ) = π1 (A, x0 )
∗
π1 (B, x0 )
...
The theorem asserts that this map is an isomorphism
...
Example
...
We want to find π1 (S n )
...
We let n = e1 ∈ S n ⊆
n+1
R
be the North pole, and s = −e1 be the South pole
...
By stereographic projection, we know that A, B ∼
= Rn
...
To do so, we can draw a cylinder S n−1 × (−1, 1), and project our A ∩ B
onto the cylinder
...
So
A∩B ∼
= S n−1 × (−1, 1) ' S n−1 , since (−1, 1) is contractible
...
Note that this works only
if S n−1 is path-connected, i
...
n ≥ 2
...
We can see this directly form the
universal property of the amalgamated free product, or note that it is the quotient
of 1 ∗ 1, which is 1
...
We have found yet another simply connected space
...
Our previous spaces were simply connected because they
were contractible
...
So this
is genuinely a new, interesting example
...
However,
the problem is that space-filling curves exist
...
Then the above proof strategy works
...
Example (RPn )
...
Now that we have proved that S n is simply connected,
we know that S n is a universal cover of RPn
...
Since p−1 (x0 ) has two elements by definition, we know that |π1 (RPn , x0 )| = 2
...
You will prove a generalization of this in example sheet 2
...
We are going to consider the operation of
wedging
...
The natural way to join them is to take the disjoint union
...
What we do is take the wedge
sum, where we take the disjoint union and then identify the base points:
X
Y
y0
x0
x0 ∼ y0
X ∧Y
Suppose we take the wedge sum of two circles S 1 ∧ S 1
...
So we take slightly more, and get the following:
x0
A
B
Each of A and B now look like this:
We see that both A and B retract to the circle
...
So π1 (A ∩ B) = 1
...
We can see that Z ∗ Z ∼
= F2 by showing that
they satisfy the same universal property
...
More generally, as long as x0 , y0 in X and Y are “reasonable”, π1 (X ∧ Y ) ∼
=
π1 (X) ∗ π1 (Y )
...
e
...
Recall that π1 (S 1 ∧ S 1 , x0 ) ∼
= F2 ∼
= ha, bi
...
Consider the map φ : F2 → Z/3 which sends a 7→ 1, b 7→ 1
...
This exists since the universal property tells us we just have to say where the
generators go, and the map exists (and is unique)
...
So there is a based covering space of S 1 ∧ S 1
˜ Let’s work out what it is
...
First, we want to know how many sheets it has, i
...
how many copies of x0
we have
...
x
˜0
˜ x
˜0 ), a does not lift to
Let’s try to lift the loop a at x
˜0
...
So it goes to another vertex
...
So we get the following
a
x
˜0
a
Since a3 ∈ ker φ, we get a loop labelled a
...
So ab−1 gives a loop
...
This is a fun game to play at home:
(i) Pick a group G (finite groups are recommended)
...
(iii) Compute the covering space corresponding to φ
...
2
The effect on π1 of attaching cells
We have started the course by talking about cell complexes, but largely ignored
them afterwards
...
The process of attaching cells is as follows: we start with a space X, get a
function S n−1 → X, and attach Dn to X by gluing along the image of f to get
X ∪f Dn :
y0
x0
0
Since we are attaching stuff, we can use the Seifert-van Kampen theorem to
analyse this
...
More precisely, the map
Theorem
...
This tells us that attaching a high-dimensional disk does not do anything to
the fundamental group
...
Again, the difficulty of applying Seifert-van Kampen theorem is that we
need to work with open sets
...
We let A = X ∪f (Dn \ {0})
...
So A deformation
˚ the interior of Dn
...
Let B = D,
˚n \ 0 ∼
A∩B =D
= S n−1 × (−1, 1)
We cannot use y0 as our basepoint, since this point is not in A ∩ B
...
Since Dn is path connected, we have a path
γ : y1
y0 , and we can use this to recover the fundamental groups based at y0
...
Since B is just a disk, and A ∩ B is simply connected (n ≥ 3 implies S n−1 is
simply connected), their fundamental groups are trivial
...
We can now use γ to change base points from y1 to y0
...
49
5
Seifert-van Kampen theorem
II Algebraic Topology
The more interesting case is when we have smaller dimensions
...
If n = 2, then the natural map π1 (X, X0 ) → π1 (X ∪f Dn , x0 ) is
surjective, and the kernel is hh[f ]ii
...
This is what we would expect, since if we attach a disk onto the loop given
by f , this loop just dies
...
As before, we get
π1 (X ∪f Dn , y1 ) ∼
= π1 (A, y1 )
∗
π1 (A∩B,y1 )
π1 (B, y1 )
...
However, π1 (A ∩ B, y1 ) ∼
= Z
...
In summary, we have
(
π1 (X)
n≥3
π1 (X ∪f D ) =
π1 (X)/hhf ii n = 2
n
This is a useful result, since this is how we build up cell complexes
...
Moreover, whenever X
is a cell complex, we should be able to write down the presentation of π1 (X)
...
Let X be the 2-torus
...
So we have our torus as a cell complex
...
We start with X (0)
...
So its fundamental group is
π1 (X (0) ) = 1
...
Finally, to get π1 (X), we have to quotient out by the boundary of our square,
which is just aba−1 b−1
...
We have the last congruence since we have two generators, and then we make
them commute by quotienting the commutator out
...
Corollary
...
There really isn’t anything that requires that finiteness in this proof, but
finiteness makes us feel more comfortable
...
Let S = {a1 , · · · , am } and R = {r1 , · · · , rn }
...
We then get
our 2-cells e2j for j = 1, · · · , n, and attaching them to X (1) by fi : S 1 → X (1)
given by a based loop representing ri ∈ F (S)
...
5
...
We first start with a definition
...
A subset A ⊆ X is a neighbourhood deformation retract if there is an open set A ⊆ U ⊆ X such that A is
a strong deformation retract of U , i
...
there exists a retraction r : U → A and
r ' idU rel A
...
Example
...
51
5
Seifert-van Kampen theorem
II Algebraic Topology
Theorem
...
Suppose that A,
B and A ∩ B are path connected, and A ∩ B is a neighbourhood deformation
retract of A and B
...
π1 (X, x0 ) = π1 (A, x0 )
A
∗
π1 (A∩B,x0 )
π1 (B, x0 )
...
Proof
...
Let U be such that U retracts
to A ∩ B
...
Let A0 = A∪V and B 0 = B ∪U
...
Since U and V retract to A ∩ B, we know A0 ' A and B 0 ' B
...
In particular, it is path connected
...
This is basically what we’ve done all the time when we enlarge our A and B
to become open
...
Let X = S 1 ∧ S 1 , the rose with two petals
...
x0
A
B
Then since {x0 } = A ∩ B is a neighbourhood deformation retract of A and B,
we know that
π1 X ∼
= π1 S 1 ∗ π1 S 1
...
4
The fundamental group of all surfaces
We have found that the torus has fundamental group Z2 , but we already knew
this, since the torus is just S 1 × S 1 , and the fundamental group of a product is
the product of the fundamental groups, as you have shown in the example sheet
...
We look at all surfaces
...
It is surprisingly difficult to get
mathematicians to agree on how we can define a surface
...
A surface is a Hausdorff topological space such that every
point has a neighbourhood U that is homeomorphic to R2
...
Needless to say, the actual surfaces will also
be different and have different homotopy groups
...
To find the fundamental group of all surfaces, we rely on the following theorem
that tells us what surfaces there are
...
If X is a compact surface, then
X is homeomorphic to a space in one of the following two families:
(i) The orientable surface of genus g, Σg includes the following (please excuse
my drawing skills):
A more formal definition of this family is the following: we start with the
2-sphere, and remove a few discs from it to get S 2 \ ∪gi=1 D2
...
(ii) The non-orientable surface of genus n, En = {RP2 , K, · · · } (where K is
the Klein bottle)
...
It would be nice to be able to compute fundamental groups of these guys
...
Example
...
This would give two tori with a hole, where the boundary of the holes are just the
dashed line
...
In general, to produce Σg , we produce a polygon with 4g sides
...
We do we care? The classification theorem tells us that each surface is homeomorphic to some of these orientable and non-orientable surfaces, but it doesn’t tell us
there is no overlap
...
However, this result lets us know that all these orientable surfaces are
genuinely different
...
We can
take the abelianization of the group π1 Σg , where we further quotient by all
commutators
...
These are clearly distinct for different values of g
...
Moreover, they are not even homotopy equivalent
...
54
6
Simplicial complexes
6
II Algebraic Topology
Simplicial complexes
So far, we have taken a space X, and assigned some things to it
...
It was easy to calculate and understand
...
What are they good for? A question we motivated
ourselves with was to prove Rm ∼
= Rn implies n = m
...
If R = R, then we would have Rm \ {0} ∼
= R \ {0} ' S 0
...
This is just a fancy
way of saying that R \ {0} is disconnected while Rm \ {0} is not for m 6= 1
...
If Rm ∼
= R2 , then we
m
2
1
1 ∼
∼
have R \ {0} = R \ {0} ' S
...
The obvious thing to do is to create some πn (X)
...
As we
would expect, this only gets harder as we get to higher dimensions
...
The problem is that πn works with groups, and groups are hard
...
We want to do some easier algebra, and a good
choice is linear algebra
...
In the rest of the course, we
will have things like H0 (X) and H1 (X), instead of πn , which are more closely
related to linear algebra
...
This sounds easy, except that we
don’t understand presentations
...
So even though we can compute the fundamental group in terms of presentations, this is not necessarily helpful
...
Computers can do it
...
This is where homology theory comes in
...
6
...
In this course, we
will be using simplicial homology, which is relatively more intuitive and can be
computed directly
...
This is not a very serious restriction
per se, since many spaces like spheres are indeed simplicial complexes
...
We now start by defining simplicial complexes, and developing some general
theory of simplicial complexes that will become useful later on
...
A finite set of points {a1 , · · · , an } ⊆ Rm is
55
6
Simplicial complexes
II Algebraic Topology
affinely independent iff
n
X
ti ai = 0 with
i=1
n
X
ti = 0 ⇔ ti = 0 for all i
...
When n = 3, the following points are affinely independent:
The following are not:
The proper way of understanding this definition is via the following lemma:
Lemma
...
Alternatively, n + 1 affinely independent points span an n-dimensional thing
...
Suppose a0 , · · · , an are affinely independent
...
i=1
Then we can rewrite this as
−
n
X
!
λi
a0 + λ1 a1 + · · · + λn an = 0
...
So affine independence implies that all
coefficients are 0
...
On the other hand, suppose a1 − a0 , · · · , an − a0 are linearly independent
...
i=0
i=0
Then we can write
t0 = −
n
X
ti
...
i=1
i=1
So linear independence implies all ti = 0
...
Definition (n-simplex)
...
e
...
i=0
i=0
The points a0 , · · · , an are the vertices, and are said to span σ
...
Example
...
Unlike arbitrary subspaces of Rn , they can be specified by a finite amount of
data
...
Definition (Face, boundary and interior)
...
The boundary is the union of
the proper faces, and the interior is the complement of the boundary
...
In particular, the interior of a vertex is the vertex itself
...
Example
...
For example, when n = 2, we get the following:
57
6
Simplicial complexes
II Algebraic Topology
We will now glue simplices together to build complexes, or simplicial complexes
...
A (geometric) simplicial complex is a finite set K of simplices in
Rn such that
(i) If σ ∈ K and τ is a face of σ, then τ ∈ K
...
Definition (Vertices)
...
Example
...
It is not a subspace of Rn
...
The polyhedron defined by K is the union of the
simplices in K, and denoted by |K|
...
Definition (Dimension and skeleton)
...
The d-skeleton K (d) of K is the union of the
n-simplices in K for n ≤ d
...
Usually, when we are given a space, say S n , it is not defined to be a simplicial
complex
...
Definition (Triangulation)
...
Example
...
The boundary ∂σ is homeomorphic
to S n−1 (e
...
the boundary of a (solid) triangle is the boundary of the triangle,
which is also a circle)
...
We can also triangulate our S n in a different way:
Example
...
So we have 2n+1 simplices in total
...
The nice thing about this triangulation is that the simplicial complex is
invariant under the antipodal map
...
As always, we don’t just look at objects themselves, but also maps between
them
...
A simplicial map f : K → L is a function
f : VK → VL such that if ha0 , · · · , an i is a simplex in K, then {f (a0 ), · · · , f (an )}
spans a simplex of L
...
So we can completely
specify a simplicial map by writing down a finite amount of information
...
In particular, they are allowed to have repeats
...
Suppose we have the standard 2-simplex K as follows:
a2
a0
a1
The following does not define a simplicial map because ha1 , a2 i is a simplex in
K, but {f (a1 ), f (a2 )} does not span a simplex:
f (a0 ), f (a1 )
f (a2 )
On the other hand, the following is a simplicial map, because now {f (a1 ), f (a2 )}
spans a simplex, and note that {f (a0 ), f (a1 ), f (a2 )} also spans a 1-simplex
because we are treating the collection of three vertices as a set, and not a
simplex
...
Lemma
...
As we said, simplicial maps are nice, but they are not exactly what we want
...
Lemma
...
There is an obvious way to define this map
...
Proof
...
i=0
i=0
The result is in L because {f (ai )} spans a simplex
...
This is clearly
continuous on σ, and is hence continuous on |K| by the gluing lemma
...
60
6
Simplicial complexes
6
...
So given a
continuous map f , we would like to find a related simplicial map g
...
The
definition we will write down is slightly awkward, but it turns out this is the
most useful definition
...
Let x ∈ |K|
...
e
...
x∈σ∈K
The link of x, written LkK (x), is the union of all those simplices that do not
contain x, but are faces of a simplex that does contain x
...
Let f : |K| → |L| be a continuous map
between the polyhedra
...
(∗)
The following lemma tells us why this is a good definition:
Lemma
...
Furthermore, if f is already simplicial on some subcomplex M ⊆ K, then we get
g|M = f |M , and the homotopy can be made rel M
...
First we want to check g is really a simplicial map if it satisfies (∗)
...
We want to show that {g(a0 ), · · · , g(an )}
spans a simplex in L
...
Since σ contains each ai , we know that x ∈ StK (ai )
for all i
...
i=0
Hence we know that there is one simplex, say, τ that contains all g(ai ) whose
interior contains f (x)
...
So they span a face of τ , as required
...
We let H : |K| × I → |L| ⊆ Rm be
defined by
(x, t) 7→ t|g|(x) + (1 − t)f (x)
...
So we need to check that im H ⊆ |L|
...
It thus follows that
H(x × I) ⊆ τ ⊆ |L|
...
Then the homotopy is rel M by
the construction above
...
Since f (v) is a vertex
and g(v) is the only vertex in StL (g(v)), we must have f (v) = g(v)
...
61
6
Simplicial complexes
II Algebraic Topology
What’s the best thing we might hope for at this point? It would be great if
every map were homotopic to a simplicial map
...
Let’s consider the following K:
How many homotopy classes of continuous maps are there K → K? Countably
many, one for each winding number
...
The problem is that we don’t have enough vertices to realize all
those interesting maps
...
Suppose
we have the following simplex:
What we do is to add a point in the center of each simplex, and join them up:
This is known as the barycentric subdivision
...
We will show that for any map, as long
as we are willing to barycentrically subdivide the simplex many times, we can
find a simplicial approximation to it
...
The barycenter of σ = ha0 , · · · , an i is
σ
ˆ=
n
X
i=0
1
ai
...
The (first) barycentric subdivision K 0 of
K is the simplicial complex:
K 0 = {hˆ
σ0 , · · · , σ
ˆn i : σi ∈ K and σ0 < σ1 < · · · < σn }
...
The rth barycentric subdivision K (r) is defined inductively as the barycentric
subdivision of the r − 1th barycentric subdivision, i
...
K (r) = (K (r−1) )0
...
|K| = |K|0 and K 0 really is a simplicial complex
...
Too boring to be included in lectures
...
Even though |K 0 | and |K| are equal, the
identity map from |K 0 | to |K| is not a simplicial map
...
e
...
Then we
can define g : K 0 → K by sending σ
ˆ 7→ vσ
...
The key theorem is that as long as we are willing to perform barycentric
subdivisions, then we can always find a simplicial approximation
...
Le K and L be simplicial complexes, and f : |K| → |L| a continuous map
...
Furthermore, if f is already simplicial on M ⊆ K, then we can choose g such
that |g||M = f |M
...
To do this, we want to quantify how “fine” our subdivisions are
...
Let K be a simplicial complex
...
We have the following lemma that tells us how large our mesh is:
Lemma
...
The key point is that as r → ∞, the mesh goes to zero
...
The proof is purely technical
and omitted
...
Proof of simplicial approximation theorem
...
We have a natural cover of |L|, namely the open stars of all vertices
...
The idea is to barycentrically subdivide our K such that each open star of K is
contained in one of these things
...
By the previous lemma, there is an r such that µ(K (r) ) < δ
...
Hence it follows that
StK (r) (x) ⊆ Bδ (x) for all vertices x ∈ VK (r)
...
Therefore defining g(x) = w, we get
f (StK (r) (x)) ⊆ StL (g(x))
...
The last part follows from the observation that if f is a simplicial map, then
it maps vertices to vertices
...
64
7
Simplicial homology
7
7
...
Instead, we are going to use this framework to move on and define some new
invariants of simplicial complexes K, known as Hn (K)
...
The drawback, however, is that the definitions are slightly less intuitive
at first sight
...
Instead, we will be using abelian groups, which really should be
thought of as Z-modules
...
e
...
Using Q makes some of our work easier, but as a result we would have lost some
information
...
Recall that we defined an n-simplex as the collection of n + 1 vertices that
span a simplex
...
What we want to do is to remember the orientation of the simplex
...
Definition (Oriented n-simplex)
...
We often denote an oriented simplex as σ, and then σ
¯ denotes the same
simplex with the opposite orientation
...
As oriented 2-simplices, (v0 , v1 , v2 ) and (v1 , v2 , v0 ) are equal, but
they are different from (v2 , v1 , v0 )
...
This is substantially harder in higher dimensions, and often we
just work with the definition instead
...
Let K be a simplicial complex
...
For each i, choose an
orientation on σi
...
This choice is not important, but we need to make it
...
Now let Cn (K) be the free abelian group with basis {σ1 , · · · , σ` }, i
...
Cn (K) ∼
=
Z`
...
In other words, an element of Cn (K) is just a formal sum of n-simplices
...
This will save us from
making exceptions for n = 0 cases later
...
In this definition, we have to choose a particular orientation for each of our
simplices
...
Note that if there are no n-simplices (e
...
when n = −1), we can still
meaningfully talk about Cn (K), but it’s just 0
...
We can think of elements in the chain group C1 (X) as “paths” in
X
...
Of course, with this setup, we can do more random things, like adding 57 copies
of σ6 to it, and this is also allowed
...
When defining fundamental groups, we had homotopies that allowed us to
“move around”
...
To do so, we need to define the boundary
homomorphisms
...
We define boundary homomorphisms
dn : Cn (K) → Cn−1 (K)
by
(a0 , · · · , an ) 7→
n
X
(−1)i (a0 , · · · , a
ˆi , · · · , an ),
i=0
where (a0 , · · · , a
ˆi , · · · , an ) = (a0 , · · · , ai−1 , ai+1 , · · · , an ) is the simplex with ai
removed
...
This is clear if
we draw some pictures in low dimensions:
66
7
Simplicial homology
II Algebraic Topology
v0
−v0
v1
v1
If we take a triangle, we get
v1
v1
v0
v2
v0
v2
An important property of the boundary map is that the boundary of a boundary
is empty:
Lemma
...
In other words, im dn+1 ⊆ ker dn
...
This just involves expanding the definition and working through the
mess
...
The nth simplicial homology
group Hn (K) is defined as
Hn (K) =
ker dn
...
Since we
are going to draw pictures, we are going to start with the easy case of k = 1
...
First, we give
these things names
...
The elements of Ck (K) are called
k-chains of K, those of ker dk are called k-cycles of K, and those of im dk+1 are
called k-boundaries of K
...
In other words, dc = 0
...
e
...
For example, if we have the following cycle:
e1
e0
e2
We have
c = (e0 , e1 ) + (e1 , e2 ) + (e2 , e0 )
...
67
7
Simplicial homology
II Algebraic Topology
So this c is indeed a cycle
...
e
...
This is why we call this a 1-boundary
...
v1
v0
v2
We see that a cycle that has a boundary has been “filled in”
...
Hence we define the homology
group as the cycles quotiented by the boundaries, and we interpret its elements
as k-dimensional “holes”
...
Let K be the standard simplicial 1-sphere, i
...
we have the following
in R3
...
Our chain groups are
C0 (K) = h(e0 ), (e1 ), (e2 )i ∼
= Z3
C1 (K) = h(e0 , e1 ), (e1 , e2 ), (e2 , e0 )i ∼
= Z3
...
Note that our notation is slightly confusing
here, since the brackets h · i can mean the simplex spanned by the vertices, or
the group generated by certain elements
...
Hence, the only non-zero boundary map is
d1 : C1 (K) → C0 (K)
...
−1 0
1
1 −1 0
0
1 −1
68
7
Simplicial homology
II Algebraic Topology
We have now everything we need to know about the homology groups, and we
just need to do some linear algebra to figure out the image and kernel, and thus
the homology groups
...
=
=
im(d1 : C1 (K) → C0 (K))
im d1
im d1
After doing some row operations with our matrix, we see that the image of d1 is
a two-dimensional subspace generated by the image of two of the edges
...
What does this H0 (K) represent? We initially said that Hk (K) should represent
the k-dimensional holes, but when k = 0, this is simpler
...
We interpret this to mean K has one
path component
...
Similarly, we have
ker d1 ∼
H1 (K) =
= ker d1
...
So we also have
H1 (K) ∼
= Z
...
The fact that H1 (K) is non-trivial means that we do indeed have a hole in the
middle of the circle
...
Let L be the standard 2-simplex (and all its faces) in R3
...
69
7
Simplicial homology
II Algebraic Topology
Since d1 is the same as before, the only new interesting boundary map is d2
...
We know that H0 (L) depends only on d0 and d1 , which are the same as for K
...
Again, the interpretation of this is that L is path-connected
...
im d2
h(e0 , e1 ) + (e1 , e2 ) + (e2 , e0 )i
This precisely illustrates the fact that the “hole” is now filled in L
...
im d3
This is zero since there aren’t any two-dimensional holes in L
...
We are now going to spend a lot of time developing formalism
...
2
Some homological algebra
We will develop some formalism to help us compute homology groups Hk (K) in
lots of examples
...
Definition (Chain complex and differentials)
...
We call these maps the differentials of C·
...
Definition (Cycles and boundaries)
...
The space of n-boundaries is
Bn (C) = im dn+1
...
The n-th homology group of C· is defined to be
Hn (C) =
ker dn
Zn (C)
=
...
Suppose we have two chain
complexes
...
In general, we want to have maps between these two sequences
...
70
7
Simplicial homology
II Algebraic Topology
Definition (Chain map)
...
In other words, the following diagram commutes:
0
d0
C0
d1
f0
0
d0
d2
C1
f1
D0
d1
C2
d3
···
f2
D1
d2
D2
d3
···
We want to have homotopies
...
How can we define homotopies for
chain complexes? It turns out we can have a completely algebraic definition for
chain homotopies
...
A chain homotopy between chain maps f· , g· :
C· → D· is a sequence of homomorphisms hn : Cn → Dn+1 such that
gn − fn = dn+1 ◦ hn + hn−1 ◦ dn
...
The arrows can be put in the following diagram, which is not commutative:
Cn−1
hn−1
dn
Cn
gn
fn
Dn
hn
dn+1
Dn+1
The intuition behind this definition is as follows: suppose C· = C· (K) and
D· = C· (L) for K, L simplicial complexes, and f· and g· are “induced” by
simplicial maps f, g : K → L
...
We further suppose
that H actually comes from a simplicial map K × I → L (we’ll skim over the
technical issue of how we can make K × I a simplicial complex
...
Let σ be an n-simplex of K, and here is a picture of H(σ × I):
Let
hn (σ) = H(σ × I)
...
What is its boundary? We’ve got the
vertical sides plus the top and bottom
...
How about the sides?
They are what we get when we pull the boundary ∂σ up with the homotopy,
i
...
H(∂σ × I) = hn−1 ◦ dn (σ)
...
Rearranging and dropping the σs, we get
dn+1 ◦ hn − hn−1 ◦ dn = gn − fn
...
So in reality, we will have to fiddle with the sign of hn a bit
to get it right, but you get the idea
...
A chain map f· : C· → D· induces a homomorphism:
f∗ : Hn (C) → Hn (D)
[c] 7→ [f (c)]
Furthermore, if f· and g· are chain homotopic, then f∗ = g∗
...
Since the homology groups are defined as the cycles quotiented by the
boundaries, to show that f∗ defines a homomorphism, we need to show f sends
cycles to cycles and boundaries to boundaries
...
If dn (σ) = 0,
then
dn (fn (σ)) = fn (dn (σ)) = fn (0) = 0
...
Similarly, if σ is a boundary, say σ = dn (τ ), then
fn (σ) = fn (dn (τ )) = dn (fn (τ ))
...
It thus follows that f∗ is well-defined
...
For any c ∈ Zn (C),
we have
gn (c) − fn (c) = dn+1 ◦ hn (c) + hn−1 ◦ dn (c)
...
So
gn (c) − fn (c) = dn+1 ◦ hn (c) ∈ Bn (D)
...
So [gn (c)] − [fn (c)] = 0 in Hn (D),
i
...
f∗ (c) = g∗ (c)
...
(i) Being chain-homotopic is an equivalence relation of chain maps
...
72
7
Simplicial homology
II Algebraic Topology
(iii) If f : C· → D· and g : D· → A· are chain maps, then
g∗ ◦ f∗ = (f· ◦ g· )∗
...
·
The last two statements can be summarized in fancy language by saying that
Hn is a functor
...
Chain complexes C· and D· are
chain homotopy equivalent if there exist f· : C· → D· and g· : D· → C· such
that
f· ◦ g· ' idD , g· ◦ f· ' idC
...
The chain complexes themselves are not necessarily the
same, but the induced homology groups will be
...
Let f· : C· → D· be a chain homotopy equivalence, then f∗ : Hn (C) →
Hn (D) is an isomorphism for all n
...
Let g· be the homotopy inverse
...
Similarly, g∗ ◦ f∗ = idH∗ (C)
...
7
...
Here, K is always a simplicial complex, and C· = C· (K)
...
Let f : K → L be a simplicial map
...
Hence it also induces f∗ : Hn (K) → Hn (L)
...
This is fairly obvious, except that simplicial maps are allowed to “squash”
simplices, so f might send an n-simplex to an (n − 1)-simplex, which is not in
Dn (L)
...
Let σ be an oriented n-simplex in K, corresponding to a basis element of
Cn (K)
...
0
f (σ) is a k-simplex for k < n
More precisely, if σ = (a0 , · · · , an ), then
(
(f (a0 ), · · · , f (an )) f (a0 ), · · · , f (an ) spans an n-simplex
fn (σ) =
...
It is immediate from this that this satisfies the chain map condition, i
...
f·
commutes with the boundary operators
...
A simplicial complex is a cone if, for some v0 ∈ Vk ,
|K| = StK (v0 ) ∪ LkK (v0 )
...
This is what the next lemma tells us
...
If K is a cone with cone point v0 , then inclusion i : {v0 } → |K| induces
a chain homotopy equivalence i· : Cn ({v0 }) → Cn (K)
...
It is clear that r· ◦ i· = id, and we need to show that i· ◦ r· ' id
...
Details are left to
the reader
...
If ∆n is the standard n-simplex, and L consists of ∆n and all its
faces, then
(
Z k=0
Hk (L) =
0 k>0
Proof
...
What we would really like is an example of non-trivial homology groups
...
An obvious candidate is the standard n-sphere
...
Let K be the standard (n − 1)-sphere (i
...
the proper faces of L
from above)
...
Z k =n−1
Proof
...
C0 (K)
C1 (L)
···
dL
n−1
dK
1
C1 (K)
Cn−1 (L)
dL
n
Cn (L)
=
0
dL
1
=
C0 (L)
=
0
···
74
dK
n−1
Cn−1 (K)
Cn (K) = 0
7
Simplicial homology
II Algebraic Topology
For k < n − 1, we have Ck (K) = Ck (L) and Ck+1 (K) = Ck+1 (L)
...
So
Hk (K) = Hk (L) = 0
...
We get the last equality since
ker dL
n−1
= Hn−1 (L) = 0
...
So
Cn (L) ∼
= Z
...
So
∼
Hn−1 (K) ∼
= im dL
n = Z
...
We say this is just a “suggestion”, since the simplicial
homology is defined for simplicial complexes, and we are not guaranteed that if
we put a different simplicial complex on S n−1 , we will get the same homology
groups
...
But this will take some time
...
H0 (K) ∼
= Zd , where d is the number of path
components of K
...
Let K be our simplicial complex and v, w ∈ Vk
...
The requirement that d1 c = w − v is equivalent to saying
c is a path from v to w
...
Most of the time, we only care about path-connected spaces, so H0 (K) ∼
= Z
...
4
Mayer-Vietoris sequence
The Mayer-Vietoris theorem is exactly like the Seifert-van Kampen theorem
for fundamental groups, which tells us what happens when we glue two spaces
together
...
M
N
75
7
Simplicial homology
II Algebraic Topology
We will learn how to compute the homology of the union M ∪ N in terms of
those of M, N and M ∩ N
...
The situation is somewhat similar here
...
The objects we need are known as
exact sequences
...
A pair of homomorphisms of abelian groups
f
A
g
B
C
is exact (at B) if
im f = ker g
...
We say it is exact if it is exact at every Ai
...
When we defined the
chain complexes, we had d2 = 0, i
...
im d ⊆ ker d
...
Algebraically, we can think of an exact sequence as chain complexes with
trivial homology groups
...
There is a particular type of exact sequences that is important
...
A short exact sequence is an exact sequence
of the form
f
g
0
A
B
C
0
What does this mean?
– The kernel of f is equal to the image of the zero map, i
...
{0}
...
– The image of g is the kernel of the zero map, which is everything
...
– im f = ker g
...
Definition (Short exact sequence of chain complexes)
...
Note that by requiring the maps to be chain maps, we imply that ik and jk
commute with the boundary maps of the chain complexes
...
If we have a short exact sequence of complexes
0
A·
i·
B·
j·
C·
0
then a miracle happens to their homology groups
...
e
...
Having an exact sequence is good, since if we know most of the terms in an
exact sequence, we can figure out the remaining ones
...
Yet, we still need to know some of
them, and since they are defined in the proof, you need to remember the proof
...
g
...
Then everything boils down to the
rank-nullity theorem
...
If, at any point, homology groups
confuse you, then you can try to work with homology groups over Q and get a
feel for what homology groups are like, since this is easier
...
Theorem (Mayer-Vietoris theorem)
...
We have the following inclusion maps:
L
i
j
N
M
k
`
K
...
Note that unlike the Seifert-van Kampen theorem, this does not require the
intersection L = M ∩ N to be (path) connected
...
However, homology groups don’t have these problems
...
All we have to do is to produce a short exact sequence of complexes
...
It is easy to see that this is a short exact sequence of chain complexes
...
It is also easy to see that in + jn is injective and kn − `n is
surjective
...
However, this is actually
often a good thing, since we can often use this to deduce the higher homology
groups from the lower homology groups
...
To do so, we need to prove the snake lemma
...
If we have a short exact sequence of complexes
0
A·
i·
B·
j·
C·
0
then there is a long exact sequence
···
Hn (A)
i∗
Hn (B)
j∗
Hn (C)
∂∗
Hn−1 (A)
i∗
Hn−1 (B)
j∗
Hn−1 (C)
···
where i∗ and j∗ are induced by i· and j· , and ∂∗ is a map we will define in the
proof
...
The idea of the proof is as follows — in the short exact sequence, we can think
of A as a subgroup of B, and C as the quotient B/A, by the first isomorphism
theorem
...
We
apply the boundary map to this representative, and then exactness shows that
this must come from some element of A
...
e
...
Proof
...
It just involves a lot of checking
of the details, such as making sure the homomorphisms are well-defined, are
actually homomorphisms, are exact at all the places etc
...
First we look at the following commutative diagram:
0
in
An
dn
0
An−1
Bn
jn
Cn
dn
in−1
Bn−1
0
dn
jn−1
Cn−1
0
To construct ∂∗ : Hn (C) → Hn−1 (A), let [x] ∈ Hn (C) be a class represented
by x ∈ Zn (C)
...
By exactness, we know the
map jn : Bn → Cn is surjective
...
Since our target is An−1 , we want to move down to the next level
...
We would be done if dn (y) is in the image of in−1
...
Since the diagram is
commutative, we know
jn−1 ◦ dn (y) = dn ◦ jn (y) = dn (x) = 0,
using the fact that x is a cycle
...
Moreover, by
exactness again, in−1 is injective
...
We have now produced our z
...
We have ∂∗ [x] = [z] as our candidate definition, but we
need to check many things:
(i) We need to make sure ∂∗ is indeed a homomorphism
...
e
...
(iv) We need to check the exactness of the resulting sequence
...
79
7
Simplicial homology
II Algebraic Topology
(ii) We check dn−1 (z) = 0
...
0
in
An
dn
0
Cn
dn
in−1
An−1
An−2
Cn−1
dn−1
in−2
0
dn
jn−1
Bn−1
dn−1
0
jn
Bn
dn−1
jn−2
Bn−2
0
Cn−2
0
We want to check that dn−1 (z) = 0
...
In particular, we know
in−2 ◦ dn−1 (z) = dn−1 ◦ in−1 (z) = dn−1 ◦ dn (y) = 0
...
So we must have
dn−1 (z) = 0
...
Then jn (y 0 − y) = 0
...
Let
a ∈ An be such that in (a) = y 0 − y
...
Hence when we pull back dn (y 0 ) and dn (y) to An−1 , the results differ
by the boundary dn (a), and hence produce the same homology class
...
We want to show that ∂∗ [x] = ∂∗ [x0 ]
...
0
in+1
An+1
dn+1
0
dn
0
An−1
Cn+1
dn+1
in
An
jn+1
Bn+1
Bn
dn+1
jn
Cn
dn
in−1
Bn−1
0
0
dn
jn−1
Cn−1
0
By definition, since [x0 ] = [x], there is some c ∈ Cn+1 such that
x0 = x + dn+1 (c)
...
By commutativity of the squares, we know
x0 = x + jn ◦ dn+1 (b)
...
Then
jn (y + dn+1 (b)) = x0
...
So dn (y) = dn (y 0 ), and
hence ∂∗ [x] = ∂∗ [x0 ]
...
When reading
this, it is helpful to look at a diagram and see how the elements are chased
along
...
(a) im i∗ ⊆ ker j∗ : This follows from the assumption that in ◦ jn = 0
...
Suppose j∗ ([b]) = 0
...
By surjectivity of jn+1 ,
there is some b0 ∈ Bn+1 such that jn+1 (b0 ) = c
...
e
...
By exactness of the sequence, we know there is some a ∈ An such
that
in (a) = b − dn+1 (b0 )
...
Since in−1 is injective, it follows that
dn (a) = 0
...
Then
i∗ ([a]) = [b] − [dn+1 (b0 )] = [b]
...
(c) im j∗ ⊆ ker ∂∗ : Let [b] ∈ Hn (B)
...
Then we compute dn (b) and then pull it
back to An+1
...
So
∂∗ (j∗ ([b])) = 0, i
...
∂∗ ◦ j∗ = 0
...
Let b ∈ Bn
be such that jn (b) = c, and a ∈ An−1 such that in−1 (a) = dn (b)
...
So we know a is a boundary,
say a = dn (a0 ) for some a0 ∈ An
...
In other words,
dn (b − in (a0 )) = 0
...
Moreover,
j∗ ([b − in (a0 )]) = [jn (b) − jn ◦ in (a0 )] = [c]
...
(e) im ∂∗ ⊆ ker i∗ : Let [c] ∈ Hn (C)
...
Then ∂∗ ([c]) = [a]
...
So i∗ ◦ ∂∗ = 0
...
So we can
find some b ∈ Bn+1 such that in (a) = dn+1 (b)
...
Then
dn+1 (c) = dn+1 ◦ jn+1 (b) = jn ◦ dn+1 (b) = jn ◦ in (a) = 0
...
Then [a] = ∂∗ ([c]) by definition of ∂∗
...
81
7
Simplicial homology
7
...
The goal is
to see that the homology groups H∗ (K) depend only on the polyhedron, and not
the simplicial structure on it
...
Note that this is a lot to prove
...
We only know simplicial maps do
...
Definition (Contiguous maps)
...
e
...
f
τ
σ
g
The significance of this definition comes in two parts: simplicial approximations of the same map are contiguous, and contiguous maps induce the same
maps on homology
...
If f, g : K → L are simplicial approximations to the same map F ,
then f and g are contiguous
...
Let σ ∈ K, and pick some s ∈ ˚
σ
...
Then the definition of simplicial approximation implies that for any simplicial
approximation f to F , f (σ) spans a face of τ
...
If f, g : K → L are continguous simplicial maps, then
f∗ = g∗ : Hn (K) → Hn (L)
for all n
...
If f (σ) and g(σ) are both faces of τ ,
then we just pick the homotopy as τ
...
Proof
...
We can now check by direct computation that this is indeed a chain homotopy
...
However, we do not yet know
that continuous maps induce well-defined maps on homologies, since to produce
simplicial approximations, we needed to perform barycentric subdivision, and
we need to make sure this does not change the homology
...
It turns out this is easy to do, and we can do
it almost arbitrarily
...
Each vertex σ
ˆ ∈ K 0 is a barycenter of some σ ∈ K
...
This defines a function a : VK 0 → VK
...
Moreover, every simplicial
approximation to the identity is of this form
...
Omitted
...
Proposition
...
Then the induced map a∗ :
Hn (K 0 ) → Hn (K) is an isomorphism for all n
...
We first deal with K being a simplex σ and its faces
...
Therefore
(
Z n=0
0 ∼
∼
Hn (K) = Hn (K ) =
0 n>0
So only n = 0 is (a little) interesting, but it is easy to check that a∗ is an
isomorphism in this case, since it just maps a vertex to a vertex, and all vertices
in each simplex are in the same homology class
...
So to understand their homology
groups, we use the Mayer-Vietoris sequence
...
We let L = K \ {σ} (note that L includes the boundary of σ)
...
We can do similarly for K 0 , and let L0 , S 0 , T 0 be the corresponding barycentric
subdivisions
...
By the
previous lemma, we see our construction of a gives a(L0 ) ⊆ L, a(S 0 ) ⊆ S and
a(T 0 ) ⊆ T
...
By
the five lemma, this implies the middle map is an isomorphism, where the five
lemma is as follows:
Lemma (Five lemma)
...
Proof
...
We now have everything we need to move from simplical maps to continuous
maps
...
To each continuous map f : |K| → |L|, there is an associated
map f∗ : Hn (K) → Hn (L) (for all n) given by
−1
f∗ = s∗ ◦ νK,r
,
where s : K (r) → L is a simplicial approximation to f , and νK,r : Hn (K (r) ) →
Hn (K) is the isomorphism given by composing maps Hn (K (i) ) → Hn (K (i−1) )
induced by simplical approximations to the identity
...
(ii) If g : |M | → |K| is another continuous map, then
(f ◦ g)∗ = f∗ ◦ g∗
...
If f : |K| → |L| is a homeomorphism, then f∗ : Hn (K) → Hn (L) is
an isomorphism for all n
...
Immediate from (ii) of previous proposition
...
We know now that homology groups is a property of the space
itself, not simplicial complexes
...
We’re not exactly there yet
...
What we would like to know is that they are
invariant under homotopy
...
The strategy is:
(i) Show that “small” homotopies don’t change the maps on Hn
(ii) Note that all homotopies can be decomposed into “small” homotopies
...
Let L be a simplicial complex (with |L| ⊆ Rn )
...
The idea of the proof is that if kf (x) − g(x)k is small enough, we can
barycentrically subdivide L such that we get a simplicial approximation to both
f and g
...
By the Lebesgue number lemma, there is an ε > 0 such that each ball of
radius 2ε in |L| lies in some star StL (w)
...
Now since g and f differ by at most ε, we know
g(Bδ (x)) ⊆ B2ε (y) ⊆ StL (w)
...
So for all v ∈ VK (R), we know
StK(r) (v) ⊆ Bδ (V )
...
We define
s : VK (r) → VL sending v 7→ w
...
Let f ' g : |K| → |L|
...
Proof
...
Since |K| × I is compact, we know H is uniformly
continuous
...
Then there is some δ such
that |s − t| < δ implies |H(x, s) − H(x, t)| < ε for all x ∈ |K|
...
Define fi : |K| → |L| by fi (x) = H(x, ti )
...
Hence (fi )∗ = (fi−1 )∗
...
e
...
This is good, since we know we can not only deal with spaces that are
homeomorphic to complexes, but spaces that are homotopic to complexes
...
Definition (h-triangulation and homology groups)
...
We define Hn (X) = Hn (K) for all n
...
Hn (X) is well-defined, i
...
it does not depend on the choice of K
...
Clear from previous theorem
...
We will now use them to do stuff
...
However, historically, this is not
why people studied algebraic topology, since there are other ways to prove this
is true
...
These can be thought of
as embeddings S ,→ S 3
...
The most interesting knot we know is the unknot U :
A less interesting but still interesting knot is the trefoil T
...
It obviously isn’t, but can we prove it? In general, given two
knots, is there any way we can distinguish if they are the same?
The idea is to study the fundamental groups of the knots
...
Instead, we look at the
fundamental groups of the complements
...
Staring at it hard enough, we can construct a surjection π1 (S 3 \ T ) → S3
...
So we know U and T
are genuinely different knots
...
6
Homology of spheres and applications
Lemma
...
We already did this computation for the standard (n − 1)-sphere ∂∆n ,
where ∆n is the standard n-simplex
...
Rn 6∼
= Rm for m 6= n
...
See example sheet 4
...
There is no
retraction Dn onto ∂Dn ∼
= S n−1
...
Proof
...
We first show the second part from the first
...
Then the following g : Dn → ∂Dn is a continuous retraction
...
Suppose r : Dn → ∂Dn
is a retraction, i
...
r ◦ i ' id : ∂Dn → Dn
...
Since r ◦ i is homotopic to the identity, this composition must also be the identity,
but this is clearly nonsense, since Hn−1 (S n−1 ) ∼
= Z while Hn−1 (Dn ) ∼
= 0
...
Note it is important that we can work with continuous maps directly, and
not just their simplicial approximations
...
For the next application, recall from the first example sheet that if n is odd,
then the antipodal map a : S n → S n is homotopic to the identity
...
87
7
Simplicial homology
II Algebraic Topology
By our previous calculation, we know a∗ is a map a∗ : Z → Z
...
We will now
compute a∗ and show that it is multiplication by −1 when n is even
...
The standard triangulation clearly doesn’t work
...
This triangulation works nicely with the antipodal map, since this maps a vertex
to a vertex
...
In the triangulation of S n given by vertices VK = {±e0 , ±e1 , · · · , ±en },
the element
X
x=
ε0 · · · εn (ε0 e0 , · · · , εn en )
ε∈{±1}n+1
is a cycle and generates Hn (S n )
...
By direct computation, we see that dx = 0
...
To show it
generates Hn (S n ), we note that everything in Hn (S n ) ∼
= Z is a multiple of the
generator, and since x has coefficients ±1, it cannot be a multiple of anything
else (apart from −x)
...
Now we can prove our original proposition
...
If n is even, then the antipodal map a 6' id
...
We can directly compute that a∗ x = (−1)n+1 x
...
So a 6' id
...
7
Homology of surfaces
We want to study compact surfaces and their homology groups
...
We will not
prove this fact, and just assume it to be true (it really is)
...
88
7
Simplicial homology
II Algebraic Topology
We also have non-compact versions of these, known as Fg , where we take the
above ones and cut out a hole:
We want to compute the homology groups of these Σg
...
However, there is a better way to do it
...
Then, we need to compute H∗ (Fg )
...
Recall that we produced
Σg by starting with a 4g-gon and gluing edges:
b2
a1
b1
a2
a1
b2
a2
b1
Now what is Fg ? Fg is Σg with a hole cut out of it
...
The Mayer-Vietoris sequence gives
the following
0
H2 (S 1 )
H2 (Fg−1 ) ⊕ H2 (F1 )
H2 (Σg )
H1 (S 1 )
H1 (Fg−1 ) ⊕ H1 (F1 )
H1 (Σg )
H0 (S 1 )
H0 (Fg−1 ) ⊕ H0 (F1 )
H0 (Σg )
0
We can put in the terms we already know, and get
0
H2 (Σg )
Z
Z2g
H1 (Σg )
Z
Z2
Z
0
By exactness, we know H2 (Σg ) = ker{Z → Z2g }
...
If we look at the picture, after the
deformation retraction, this loop passes through each one-cell twice, once with
each orientation
...
So H2 (Σg ) = Z
...
7
Simplicial homology
II Algebraic Topology
We now claim that Z → Z2 is injective — this is obvious, since it sends 1 7→ (1, 1)
...
So
H1 (Σg ) ∼
= Z2g
...
We write down
the long exact sequence, and put in the terms we already know
...
7
...
It turns out
we can use rational coefficients instead
...
If we use rational coefficients, since Q is a field, this becomes a
vector space, and we can use a lot of nice theorems about vector spaces, such as
the rank-nullity theorem
...
Definition (Rational homology group)
...
That is, Cn (K, Q) is the vector space over Q with basis the n-simplices of K
(with a choice of orientation)
...
Hn (K; Q) ∼
=
Bn (K; Q)
Now our homology group is a vector space, and it is much easier to work with
...
Fortunately,
the way in which we lose information is very well-understood and rather simple
...
If Hn (K) ∼
= Zk ⊕ F for F a finite group, then Hn (K; Q) ∼
= Qk
...
Exercise
...
In some cases this information loss is not very significant,
but in certain cases, it can be
...
Example
...
otherwise
7
Simplicial homology
We also have
II Algebraic Topology
Q
∼
Hn (Σg , Q) = Q2g
0
k = 0, 2
...
However, for the non-orientable surfaces, we have
k=0
Q
n−1
Hk (En , Q) = Q
,
k=1
0
otherwise
This time, this is different from the integral coefficient case, where we have an
extra Z2 term in H1
...
Also, maps on homology groups
are simply linear maps, i
...
matrices, and we can study their properties with
proper linear algebra
...
This works for two-dimensional surfaces, and we would like to
extend this to higher dimensions
...
However, if we define it
this way, it is not clear that this is a property of the space itself, and not just a
property of the triangulation
...
The Euler characteristic of a triangulable
space X is
X
χ(X) =
(−1)i dimQ Hi (X; Q)
...
We will later show this is equivalent to what we used to have
...
Definition (Lefschetz number)
...
i≥0
Why is this a generalization of the Euler characteristic? Just note that the
trace of the identity map is the number of dimensions
...
Example
...
7
Simplicial homology
II Algebraic Topology
Example
...
So
L(α) = 1 + (−1)n (−1)n+1 = 1 − 1 = 0
...
We
will soon see why this is the case
...
Before that, we want to understand the Lefschetz number first
...
However, we would like to understand the Lefschetz number in terms of
the chain groups, since these are easier to comprehend
...
Lemma
...
Let A : V → V be a linear map such that A(W ) ⊆ W
...
Then
tr(A) = tr(B) + tr(C)
...
In the right basis,
A=
A0
...
This makes our life much easier when it
comes to computation
...
Let f· : C· (K; Q) → C· (K; Q) be a chain map
...
This is a great corollary
...
However, to actually do computations, we want to work
with the chain groups and actually calculate with chain groups
...
There is an exact sequence
0
Bi (K; Q)
Zi (K; Q)
Hi (K; Q)
0
This is since Hi (K, Q) is defined as the quotient of Zi over Bi
...
Let fiH , fiB , fiZ , fiC be the various
maps induced by f on the corresponding groups
...
i≥0
Because of the alternating signs in dimension, each fiB appears twice in the sum
with opposite signs
...
i≥0
Since tr(id |Ci (K;Q) ) = dimQ Ci (K; Q), which is just the number of i-simplices,
this tells us the Euler characteristic we just defined is the usual Euler characteristic, i
...
X
χ(X) =
(−1)i number of i-simplices
...
Theorem (Lefschetz fixed point theorem)
...
If L(f ) 6= 0, then f has a fixed point
...
We prove the contrapositive
...
We will show
that L(f ) = 0
...
We know this is non-zero, since f has no fixed
point, and X is compact (and hence the infimum point is achieved by some x)
...
We now let
g : K (r) → K
be a simplicial approximation to f
...
Also, we know
|f (x) − x| ≥ δ
...
2
So we must have g(x) 6∈ σ
...
94
7
Simplicial homology
II Algebraic Topology
Now we compute L(f ) = L(|g|)
...
So we need to
compose it with si : Ci (K; Q) → Ci (K (r) ; Q) induced by inverses of simplicial
approximations to the identity map
...
So gi ◦ si
takes every simplex off itself
...
Example
...
g
...
So f
has a fixed point
...
Suppose G is a path-connected topological group, i
...
X is a group
and a topological space, and inverse and multiplication are continuous maps
...
This implies L(rg ) = 0
...
So
χ(G) = L(idG ) = L(rg ) = 0
...
This is quite fun
...
Which can be
topological groups? The torus is, since it is just S 1 × S 1 , and S 1 is a topological
group
...
However, other surfaces cannot since they
don’t have Euler characteristic 0
Title: Mathematics
Description: This is a 91-page simplified version of algebraic topology. It's concise and self-explanatory and can used to get in-depth understanding without the need of further assistance. The lecture was delivered by H. Wilton in 2015 at Cambridge University. It is organized such that all sub-topics are tabulated on a table of contents for easy navigation. It is expected that students who use this notes must not have problems with algebraic topology going forward.
Description: This is a 91-page simplified version of algebraic topology. It's concise and self-explanatory and can used to get in-depth understanding without the need of further assistance. The lecture was delivered by H. Wilton in 2015 at Cambridge University. It is organized such that all sub-topics are tabulated on a table of contents for easy navigation. It is expected that students who use this notes must not have problems with algebraic topology going forward.