(0.00) In this section, we're going to introduce the idea of
continuous paths, loops and homotopies of loops. I will assume you're
happy with the notions of topological space and continuous
maps between topological spaces. If you haven't come across these
ideas, I have a series of videos covering these ideas which you can
watch at any point during the module. If you like, you can mentally
replace the word ``topological space'' with ``metric space'' until
you've watched those videos.
Paths
(1.06) Let X be a topological space. A path in
X is a continuous map γ:[0,1]→X. You can think
of this as a continuous parametric curve γ(t) with
parameter t∈[0,1].
The map γ(t)=(t,0) is a path in the plane
X=R2 which moves along the x-axis from (0,0)
to (1,0).
The map γ(t)=(cos(2πt),sin(2πt)) is another path in
the plane which moves around the unit circle. This is a loop
(it starts and ends at the same point (1,0).
A path γ:[0,1]→X is called a loop if
γ(0)=γ(1). It is said to be a loop based at
x if γ(0)=γ(1)=x.
Free homotopy of loops
(4.06) Roughly speaking, a free homotopy of loops is a continuous
one-parameter family of loops γs(t) interpolating between
two loops γ0(t) and γ1(t). More precisely:
A free homotopy of loops is a continuous map H:[0,1]×[0,1]→X such that γs(t):=H(s,t) is a loop
for each fixed s∈[0,1], that is H(s,0)=H(s,1) for all
s∈[0,1].
Figure 1.In this figure, we see the domain and target of a
free homotopy H:[0,1]×[0,1]→X. The loops
γs(t)=H(s,t) are indicated in red for
s=0,14,12,34,1. The path traced by
basepoints is drawn in thick black; as a visual aid, the path traced
by the midpoints of the loops is also drawn in dotted black.
We will often write γs instead of H to emphasise the
fact that H is a family of loops.
Continuity of H as a map of two variables is what gives us a
continuous family of continuous loops.
Based homotopy of loops
(7.53) We will need a slightly more restricted notion of homotopy.
If x∈X is a basepoint, a homotopy based at x is a
homotopy H:[0,1]×[0,1]→X where H(s,0)=H(s,1)=x
for all s∈[0,1]. In other words, all the loops
γs(t)=H(s,t) pass through the basepoint x at
t=0,1. If we have a based homotopy γs we will write
γ0≃γ1 and say that γ0 is homotopic to
γ1.
By focusing on based loops (and based homotopy) we will be able to
concatenate loops (because they all start and end at the same point)
and this will give us a group structure on (homotopy classes of)
loops.
Loops in Rn are contractible
(11.04) If γ:[0,1]→Rn is a loop
based at the origin 0 then γ is based-homotopic to the
constant loopt↦0.
The map H(s,t)=(1−s)γ(t) is a continuous map such that
H(s,0)=0=H(s,1), so it is a based homotopy. At s=0, we get
γ0(t)=H(0,t)=γ(t). At s=1, we get
γ1(t)=H(1,t)=0, so this is a based homotopy from
γ to the constant loop.
A homotopy from a loop γ to a constant loop is called a
nullhomotopy (and we say that γ is contractible if
it is nullhomotopic).
Pre-class questions
Is the homotopy γR from the previous video a based
homotopy or a free homotopy?
Navigation
Previous video: 1.01 Winding numbers and the fundamental theorem
of algebra.
Next video: 1.03 Concatenation and the fundamental group.