CS 880: Quantum Information Processing 10/25/2010

Lecture 20: Density Operator Formalism

Instructor: Dieter van Melkebeek Scribe: Dalibor Zelen y
So far in this course we have been working in the setting where the goal is to realize a relation by
means of some computation. This involved only one party that was performing the computation.
In todays lecture and several following lectures, we will focus on systems where multiple parties
participate in the computation. We develop the density operator formalism which is suitable for
describing multiparty systems. It turns out that we can use this formalism to describe the evolution
of a quantum system, too, and that it is in some sense superior to our original way of describing
things.
1 Density Operator
We start with the denition of the density operator, give some examples, and prove some properties
of density operators. To conclude this section, we show how to represent the evolution of a quantum
system using density operators.
Denition 1 (Density operator). For a pure state |, the density operator is = | |. For a
mixed state {(p
i
, |
i
)}
i
, the density operator is =

i
p
i
| |.
When we apply the density operator to a state , we get the projection of onto , that is,
| = | |. Also note that the density operator corresponding to a mixed state is just a
convex combination of density operators for the individual pure states that form the mixed state.
1.1 Examples of Density Operators
We now present some examples of density operators. As the next two examples show, two dierent
mixed states can have the same density operator.
Example: Lets compute the density operator corresponding to {(
1
2
, |0), (
1
2
, |1)}. The density
operators for |0 and |1 are

0
= (1, 0)
T
(1, 0) =
_
1 0
0 0
_
and
1
= (0, 1)
T
(0, 1) =
_
0 0
0 1
_
.
Now we take their convex combination based on the probabilities describing our mixed state and
get that =
1
2

0
+
1
2

1
=
1
2
I.
Example: Now we compute the density operator corresponding to {(
1
2
, |+), (
1
2
, |)}. The density
operators for |+ and | are

+
=
1

2
(1, 1)
T
1

2
(1, 1) =
1
2
_
1 1
1 1
_
and

=
1

2
(1, 1)
T
1

2
(1, 1) =
_
1 1
1 1
_
.
Now we take their convex combination based on the probabilities describing our mixed state and
get that =
1
2

+
+
1
2

=
1
2
I.
1
We see that the mixed states {(
1
2
, |0), (
1
2
, |1)} and {(
1
2
, |+), (
1
2
, |)} have the same corre-
sponding density operators. As we will see later, this implies that no quantum procedure can
distinguish between these two mixed states.
We conclude our list of examples with a density operator corresponding to a two-qubit state.
Example: The density operator corresponding to
1

2
(|00 + |11) =
1

2
(1, 0, 0, 1)
T
is
1

2
(1, 0, 0, 1)
T
1

2
(1, 0, 0, 1) =
1
2
_
_
_
_
1 0 0 1
0 0 0 0
0 0 0 0
1 0 0 1
_
_
_
_
.

The state in the last example,

1

2
(|00 + |11), is called an EPR pair.
1.2 Properties of Density Operators
We use traces extensively when describing properties of density operators, so we start with some
properties of the trace of a matrix. Afterwards, we state three properties of density operators.
Recall that the trace of a matrix M, denoted Tr (M), is the sum of the diagonal entries of M,
i.e., Tr (M) =

i
M
ii
. It is easy to see that Tr (AB) = Tr (BA). Note, however, that it is not true
in general that Tr (ABC) = Tr (ACB). The equality holds only for cyclic shifts of a product of
matrices. For example, Tr (ABC) = Tr (CAB) holds.
An equivalent denition of the trace is Tr (A) =

i
where the
i
are the eigenvalues of A. To
see this for the case where A has a basis of eigenvectors, write V as the matrix whose columns are
As eigenvectors, and note that AV = V . Since V s columns form a basis for A, V is invertible,
and we have A = V V
1
where is the matrix with
ii
=
i
and with zeros in the o-diagonal
entries. Now Tr (A) = Tr
_
V V
1
_
= Tr
_
V
1
V
_
= Tr () =

i
.
Also recall that a matrix is positive semi-denite if x| M |x 0 for all x.
Claim 1. Let be a density operator. Then Tr () = 1.
Proof. First consider a pure state | =

x
|x. Then the diagonal entry
ii
of the corresponding
density operator is
2
x
, and we know that

2
x
= 1.
For a mixed state, just notice that the resulting density operator is a convex combination of
matrices with trace 1.
Claim 2. The density operator is Hermitian.
Proof. For a state

x
|x, the corresponding density operator is = (

x
|x)(

y

y
y|).
Then
xy
=
x

y
, and
yx
=
y

x
=
x
y =
yx
. This shows that is Hermitian for pure states.
For mixed states, just note that a convex combination of Hermitian matrices is Hermitian.
Claim 3. The density operator is positive semi-denite.
Proof. Consider a state |. Then | | = | (

i
p
i
|
i

i
|) | =

i
p
i
|
i

i
| =

i
p
i
| |
i
|
2
0, where the inequality follows because p
i
0 for all i.
Exercise 1. It turns out that we dont need Claim 2 because every positive semi-denite operator
is Hermitian. Prove this assertion.
2
The combination of the necessary conditions from Claim 1 and Claim 3 actually yields a sucient
condition for a matrix to be a density operator. We have the following theorem.
Theorem 1. The matrix describes a density operator if and only if Tr () = 1 and is positive
semi-denite.
Proof. We argued the forward direction in the proofs of Claims 1 and 3.
For the reverse direction, assume Tr () = 1 and is positive semi-denite. We need to nd a
mixed state whose density operator is described by . Since is positive semi-denite, its Her-
mitian, and thus has an orthonormal basis of eigenvectors, say |
1
, . . . , |
k
, with corresponding
eigenvalues
1
, . . . ,
k
. This means that we can write =

i
|
i

i
|. Since is Hermitian, it
has real eigenvalues. Since its also positive semi-denite, all eigenvalues are non-negative. Finally,
since the trace of is 1, the eigenvalues dene a probability distribution, so is the density operator
corresponding to the mixed state {(
i
, |
i
)}
i
.
1.3 Describing the Evolution of a Quantum System
Now we show that we can describe quantum computation using density operators. For that, we
need to describe the density operator

corresponding to the state |

obtained from state |

either by applying a unitary operation to | or by making a measurement of |.
Lets start with applying a unitary operation U to the state |. The new state is |

= U |,
so the corresponding density operator is

= U | (U |)

= U | | U

= UU

. We use
linearity to get the density operator in the case of mixed states.
Now suppose we make a measurement of a state | whose density operator is . We measure
with respect to some orthogonal basis {|
1
, . . . , |
k
}. The state is a linear combination of the
basis vectors, say | =

i
|
i
. We observe the state |
i
with probability |
i
|
2
, so the new
state is a mixed state {(|
i
|
2
, |
i
)}
i
, and its corresponding density operator is

i
|
i
|
2
|
i

i
| =

i
|
i

i

i

i
| (1)
Note that if we multiply on the right with |
j
and on the left with
j
|, we get the probability
that we observe
j
. This follows from the second summation in (1) because
j
|
i
= 1 if i = j,
and is zero otherwise. Thus, another way of writing (1) is

i

i
| |
i
|
i

i
|. Once again,
we can apply linearity to get the resulting density operator when we observe a mixed state.
With this in hand, we can prove the following theorem.
Theorem 2. Two states are distinguishable by some quantum process if and only if their density
operators are dierent.
Proof. Assume that two density operators are the same. We just showed in the previous paragraphs
that we only need the density operator in order to describe the outcome of some quantum process,
and gave an expression for the density operator corresponding to the next state of the system.
Thus, any quantum process operating on two states with the same density operators evolves the
same for both of the states, results in the same nal density operator for the two nal states, and,
most importantly, the probability of observing a string x is the same for both states. Thus, since
we rely on observations to decide on the output of quantum algorithms, we cannot tell from the
distribution of the observations which state we were in at the beginning.
3
Now suppose the density operators of two states are dierent. Since they are both density
operators, they have a dierent orthogonal basis of eigenvectors, or the eigenvalue corresponding
to some eigenvector is dierent for the two density operators. In either case, we get a dierent
distribution of observed basis vectors, and we can distinguish between the two states.
Exercise 2. Make the second paragraph in the proof of Theorem 2 more formal.
2 Two-Party Systems
In a two-party system, two parties, Alice and Bob, have access to two dierent parts of a quantum
register. Alice applies unitary transformations and observations to her part of the register without
aecting Bobs part, and vice versa. The general form of the state is

s,t

s,t
|s |t where the rst
component (the state |s) belongs to Alice and the second component belogs to Bob. To Alice, the
state of the system looks like a mixed state over all possible states that Bobs part of the quantum
register could be in. Thus, Alices state is
__
Pr[t],

s,t
|s
_
Pr[t]
__
t
where Pr[t] =

s
|
s,t
|
2
,
and there is a symmetric expression for Bobs state.
Lets now nd the density operator
A
for Alice. We call this the reduced density operator.

A
=

t
Pr[t]

s,t
|s
_
Pr[t]

s

s

,t
s

|
_
Pr[t]
=

s,s

s,t

,t
_
|s

s,s

t
(
AB
)
(s,t),(s

,t)
_
|s

. (2)
where the inner sum in (2) is denoted (Tr
B
(
AB
))
s,s
and is called the trace with respect to B. The
matrix
AB
in (2) is the density operator for the whole system. It follows from (2) that for states
(not superpositions) s, t, s

, and t

we have Tr
B
(|s s

| |t t

|) = t|t

|s s

|, with t|t

= 1 if
t = t

and zero otherwise, and we use linearity for superpositions. This may look a little confusing,
so lets look at an example.
Example: Suppose Alice and Bob operate on a two-qubit system, where the rst qubit belongs to
Alice and the second qubit belongs to Bob. The density operator is
=
_
_
_
_

00,00

00,01

00,10

00,11

01,00

01,01

01,10

01,11

10,00

10,01

10,10

10,11

11,00

11,01

11,10

11,11
_
_
_
_
.
Then the trace with respect to B is the matrix
Tr
B
() =
_

00,00
+
01,01

00,10
+
01,11

10,00
+
11,01

10,10
+
11,11
_
.
4
We see that (Tr
B
())
(s,s

)
is the trace of a submatrix of where Alices part of the rst index (i.e.,
the rst bit of the rst index in our case) is xed to s and Alices part of the second index is xed
to s

. Using this observation, we see that the trace with respect to A is the matrix
Tr
A
() =
_

00,00
+
10,10

00,01
+
10,11

01,00
+
11,10

01,01
+
11,11
_
.

Example: Let
=
1
2
_
_
_
_
1 0 0 1
0 0 0 0
0 0 0 0
1 0 0 1
_
_
_
_
be the density operator for the EPR pair. Then we have Tr
A
() = Tr
B
() =
1
2
I.
2.1 Schmidt Decomposition
Suppose we have a system where Alice can act on some part of it and Bob acts on the rest. The
state is described by |
AB
. We can always write this state as a linear combination of the standard
basis vectors. The next theorem states that we can do better. It is possible to write the state as a
tensor product of two linear combinations of vectors coming from two orthonormal bases, one for
Alice and one for Bob, that are potentially much smaller. Moreover, both bases have the same set
of eigenvalues.
Theorem 3. Given a state |
AB
, there exist orthonormal bases {|
i
}
i
for Alices part of the
state and {|
i
}
i
for Bobs part of the state, and
i
[0, 1] such that |
AB
=

i
|
i
|
i
with

2
i
= 1.
Before we prove Theorem 3, lets see how we can use it to obtain reduced density operators
A
and
B
for Alice and Bob. It turns out that we can use traces with respect to B and A, respec-
tively. To see that, note
AB
=

2
i
|
i
|
i

i
| |
i
=

2
i
(|
i

i
|)(|
i

i
|), and if we trace
out the B component we get
A
= Tr
B
_
i

2
i
(|
i

i
|) (|
i

i
|)
_
=

2
i

i
|
i
|
i

i
| =

2
i
|
i

i
|, where the last equality follows because
i
|
i
= 1. Similarly, we get
B
=
Tr
A
(
AB
) =

2
i
|
i

i
|.
Proof Sketch for Theorem 3. Look at a superposition

s,t

s,t
|s |t. The values
s,t
form a matrix
A = (
st
)
s,t
, which we express using singular value decomposition as A = UV where U and V
are orthogonal and is the matrix containing the singular values of A on the diagonal. We now
have
s,t
=

i
U
si

ii
V
it
, so we can use the columns of U and V as the bases for Alices and Bobs
parts of the state, respectively.
2.2 Purication
We use the Schmidt decomposition to go from a density operator representing the state of the system
to a reduced density operator correspodning to what is seen by one of the parties participating in
the computation. Our goal here is the opposite. We start with a mixed state described by the
density operator
A
and want to construct from it a pure state |
AB
of a bigger system so that

A
= Tr
B
(|
AB

AB
|). We have
A
=

i
|
i

i
|, and let |
AB
=

i
|
i
|
i
. As we
can see, we are dening Bobs part of the state to be the same as Alices part.
5
3 Next Time
In the following lectures, we will see some applications of density operators.
6