Let be a finite group. recall that a unitary representation of
is a pair
consisting of a nonzero Hilbert space
together with a group homomorphism
where
is the group of unitary operators on
We are going to spend the rest of the course studying the category whose objects are finite-dimensional unitary representations of
Morphisms in
are defined as follows.
Definition 20.1. Let and
be unitary representations of
A linear transformation
is said to intertwine the actions
and
if
Declare the set of of morphisms from
to
to be the set of intertwining maps as above.
We have now defined the category of unitary representations of
This category encodes all the ways in which
can act on finite-dimensional Hilbert spaces. It has more objects than the category
of finite-dimensional Hilbert spaces, but its Hom-spaces are smaller.
Problem 20.1. Prove that is a subspace of
Moreover, prove that if
then
Now let us discuss a few ways in which the definition of is well-chosen.
First consider the space of endomorphisms of a unitary representation
of
This is the subspace of
consisting of linear operators
on
which commute with each of the unitary operators
Equivalently, if we let be the linear representation of
corresponding to the unitary representation
via
, then
is the centralizer of the image of
in
under
Thus, since the centralizer of any subalgebra is again a subalgebra, we have that
is a subalgebra of
Isomorphisms in are easy to understand:
is nonempty if and only if
and it consists of all linear bijections
We know from Math 202A that if
is nonempty then there exists an isometry
The set
of isomorphisms between unitary representations
and
is more constrained, since it consists of linear bijections
which intertwine the actions
and
and it may be empty even if a linear bijection
exists.
Problem 20.2. Prove that is nonempty if and only if it contains an isometry
(This can be done using either singular value decomposition or polar decomposition).
Problem 20.2 gives a matrix interpretation of what it means for two unitary representations and
to be isomorphic. Namely, let
be an isometry. Choose any orthonormal basis
and let
be the corresponding orthonormal basis of
Then, we have
.
In other words, for any ordering , taking the corresponding ordering
of
we have that the
unitary matrices
and
coincide for all
So, isomorphic representations of
represent the elements of
as exactly the same unitary matrices. An obvious but important consequence of this is the following.
Theorem 20.1. Isomorphic objects in have the same character.
One of the many miracles of group representation theory is that the converse of Theorem 20.1 holds: the character of a unitary representation characterizes it up to isomorphism (hence the name). As a first step towards establishing this result, we consider another important aspect of , namely that given any two objects we can build a third as follow.
Definition 20.2. Let and
be unitary representations of
The corresponding adjoint representation
is defined by
Let us place this construction in a more general framework. Given two finite-dimensional Hilbert spaces and
equip
with the Hilbert-Schmidt scalar product.
Definition 20.3. The adjoint action of on
is the group homomorphism
defined by
Problem 20.3. Check that is indeed a unitary operator on
, and that
is indeed a group homomorphism.
Now let us calculate the matrix elements of in terms of the matrix elements of
and
To do so, let
and
be orthonormal bases, and let
be the corresponding orthonormal basis of
Proposition 20.3. We have
Proof: We have
which is the trace of an operator in We compute this trace as
Now expand
to get
which completes the proof.
Corollary 20.4. We have
Proof: Proposition 20.3 gives
as claimed.
From the above, we get the following relationship between the characters of two given unitary representations and
and the character of the corresponding adjoint representation
Corollary 20.5. For all we have