Lecture 13
Lecture 13
Unsupervised
learning
introduc3on
Machine
Learning
Supervised
learning
Training
set:
Andrew
Ng
Unsupervised
learning
Training
set:
Andrew
Ng
Applica2ons
of
clustering
Andrew
Ng
K-‐means
algorithm
T-‐shirt sizing
Weight
Height
Andrew
Ng
Clustering
Op3miza3on
objec3ve
Machine
Learning
K-‐means
op2miza2on
objec2ve
=
index
of
cluster
(1,2,…,
)
to
which
example
is
currently
assigned
=
cluster
centroid
(
)
=
cluster
centroid
of
cluster
to
which
example
has
been
assigned
Op3miza3on
objec3ve:
Andrew
Ng
K-‐means
algorithm
Repeat
{
for
=
1
to
:=
index
(from
1
to
)
of
cluster
centroid
closest
to
for
=
1
to
:=
average
(mean)
of
points
assigned
to
cluster
}
Andrew
Ng
Clustering
Random
ini3aliza3on
Machine
Learning
K-‐means
algorithm
Repeat
{
for
=
1
to
:=
index
(from
1
to
)
of
cluster
centroid
closest
to
for
=
1
to
:=
average
(mean)
of
points
assigned
to
cluster
}
Andrew
Ng
Random
ini2aliza2on
Should
have
Randomly
pick
training
examples.
Set
equal
to
these
examples.
Andrew
Ng
Local
op2ma
Andrew
Ng
Random
ini2aliza2on
For
i
=
1
to
100
{
Randomly
ini3alize
K-‐means.
Run
K-‐means.
Get
.
Compute
cost
func3on
(distor3on)
}
Andrew
Ng
Clustering
Choosing
the
number
of
clusters
Machine
Learning
What
is
the
right
value
of
K?
Andrew
Ng
Choosing
the
value
of
K
Elbow
method:
Cost
func3on
Cost
func3on
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
Andrew
Ng
Choosing
the
value
of
K
Some3mes,
you’re
running
K-‐means
to
get
clusters
to
use
for
some
later/downstream
purpose.
Evaluate
K-‐means
based
on
a
metric
for
how
well
it
performs
for
that
later
purpose.
E.g.
T-‐shirt
sizing
T-‐shirt
sizing
Weight
Weight
Height
Height
Andrew
Ng