Understanding Quantum Information and Computation

Basics of quantum information

Lesson 4: Entanglement in action

Contents

1. Quantum teleportation
2. Superdense coding
3. The CHSH game
Alice and Bob
• Alice and Bob are names given to hypothetical entities or agents in systems,
protocols, and games that involves the exchange of information.
• They are assumed to be in different locations.
• The specific roles they play must be clarified in different situations.
• Additional characters (e.g., Charlie, Diane, Eve, and Mallory) may be
introduced as needed.
Remarks on entanglement
In Lesson 2, we encountered this example of an entangled state of two qubits:

+ 1 1
∣ϕ ⟩ = √ ∣00⟩ + √ ∣11⟩
2 2

We also encountered this example of a probabilistic state of two bits:

1 1
∣00⟩ + ∣11⟩
2 2

It is typical in the study of quantum information and computation that we view

entanglement as a resource that can be used to accomplish different tasks.
+
When we do this we view the state ∣ϕ ⟩ as representing one unit of entanglement called
an e-bit.
Terminology
To say that Alice and Bob share an e-bit means that Alice has a qubit A, Bob has
+
a qubit B, and together the pair (A, B) is in the state ∣ϕ ⟩.
1. Quantum teleportation
Teleportation set-up
Scenario

Alice has a qubit Q that she wishes to transmit to Bob.

• Alice is unable to physically send Q to Bob — she is only able to
send classical information.
• Alice and Bob share an e-bit.

Remarks

• The state of Q is “unknown” to both Alice and Bob.

• Correlations (including entanglement) between Q and other systems
must be preserved by the transmission.
• The no-cloning theorem implies that if Bob receives the transmission,
Alice must no longer have the qubit in its original state.
Teleportation protocol
Circuit description

Q
∣ψ⟩ H

⎧
⎪
⎪
⎪
A
+ Alice
⎪
⎪
⎪
⎪
⎪
⎪
+ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
Bob
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ B
⎪
⎪ X Z ∣ψ⟩
⎩

Initial conditions

Alice and Bob share one e-bit: Alice has a qubit A, Bob has a qubit B,
+
and (A, B) is in the state ∣ϕ ⟩.

Alice also has a qubit Q that she wishes to transmit to Bob.

Teleportation protocol
Operation performed by Bob
Q
∣ψ⟩ H
b
1 if ab = 00
⎧
⎪
⎪
A
+ Alice
⎪
⎪
⎪
⎪
a
Z if ab = 01
⎪
⎪
⎪
+ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
Bob
X if ab = 10
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
B
X Z ∣ψ⟩
ZX if ab = 11
⎩

Protocol

1. Alice performs a controlled-NOT operation, where Q is the

control and A is the target.
2. Alice performs a Hadamard operation on Q.
3. Alice measures A and Q, obtaining binary outcomes a and b,
respectively.
4. Alice sends a and b to Bob.
5. Bob performs these two steps:
5.1 If a = 1, then Bob applies an X operation to the qubit B.
5.2 If b = 1, then Bob applies a Z operation to the qubit B.
Teleportation analysis
α∣0⟩ + β∣1⟩ H

⎧
⎪
⎪
⎪
+ Alice
⎪
⎪
⎪
⎪
⎪
⎪
+ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
Bob
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ X Z α∣0⟩ + β∣1⟩
⎩

∣π0 ⟩ ∣π1 ⟩ ∣π2 ⟩

α∣000⟩ + α∣110⟩ + β∣001⟩ + β∣111⟩

∣π0 ⟩ = √
2
α∣000⟩ + α∣110⟩ + β∣011⟩ + β∣101⟩
∣π1 ⟩ = √
2
α∣00⟩∣+⟩ + α∣11⟩∣+⟩ + β∣01⟩∣−⟩ + β∣10⟩∣−⟩
∣π2 ⟩ = √
2
Teleportation analysis
α∣0⟩ + β∣1⟩ H

⎧
⎪
⎪
⎪
+ Alice
⎪
⎪
⎪
⎪
⎪
⎪
+ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
Bob
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ X Z α∣0⟩ + β∣1⟩
⎩

∣π0 ⟩ ∣π1 ⟩ ∣π2 ⟩

α∣00⟩∣+⟩ + α∣11⟩∣+⟩ + β∣01⟩∣−⟩ + β∣10⟩∣−⟩

∣π2 ⟩ = √
2
α∣00⟩(∣0⟩ + ∣1⟩) + α∣11⟩(∣0⟩ + ∣1⟩) + β∣01⟩(∣0⟩ − ∣1⟩) + β∣10⟩(∣0⟩ − ∣1⟩)
=
2
α∣000⟩ + α∣001⟩ + α∣110⟩ + α∣111⟩ + β∣010⟩ − β∣011⟩ + β∣100⟩ − β∣101⟩
=
2
1 1 1 1
= (α∣0⟩ + β∣1⟩)∣00⟩ + (α∣0⟩ − β∣1⟩)∣01⟩ + (α∣1⟩ + β∣0⟩)∣10⟩ + (α∣1⟩ − β∣0⟩)∣11⟩
2 2 2 2
Teleportation analysis
α∣0⟩ + β∣1⟩ H

⎧
⎪
⎪
⎪
+ Alice
⎪
⎪
⎪
⎪
⎪
⎪
+ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
Bob
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ X Z α∣0⟩ + β∣1⟩
⎩

∣π0 ⟩ ∣π1 ⟩ ∣π2 ⟩

1 1 1 1
∣π2 ⟩ = (α∣0⟩ + β∣1⟩) ∣00⟩ + (α∣0⟩ − β∣1⟩) ∣01⟩ + (α∣1⟩ + β∣0⟩) ∣10⟩ + (α∣1⟩ − β∣0⟩) ∣11⟩
2 2 2 2
1 2 1
Pr(ab = 00) = ∥α∣0⟩ + β∣1⟩∥ =
4 4
1 2 1
Pr(ab = 01) = ∥α∣0⟩ − β∣1⟩∥ =
4 4
1 2 1
Pr(ab = 10) = ∥α∣1⟩ + β∣0⟩∥ =
4 4
1 2 1
Pr(ab = 11) = ∥α∣1⟩ − β∣0⟩∥ =
4 4
Teleportation analysis
α∣0⟩ + β∣1⟩ H

⎧
⎪
⎪
⎪
+ Alice
⎪
⎪
⎪
⎪
⎪
⎪
+ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
Bob
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ X Z α∣0⟩ + β∣1⟩
⎩

∣π0 ⟩ ∣π1 ⟩ ∣π2 ⟩

1 1 1 1
∣π2 ⟩ = (α∣0⟩ + β∣1⟩) ∣00⟩ + (α∣0⟩ − β∣1⟩) ∣01⟩ + (α∣1⟩ + β∣0⟩) ∣10⟩ + (α∣1⟩ − β∣0⟩) ∣11⟩
2 2 2 2

ab Probability Conditional state of (B, A, Q) Operation on B Final state of B

1
00 4
(α∣0⟩ + β∣1⟩)∣00⟩ 1 α∣0⟩ + β∣1⟩
1
01 4
(α∣0⟩ − β∣1⟩)∣01⟩ Z α∣0⟩ + β∣1⟩
1
10 4
(α∣1⟩ + β∣0⟩)∣10⟩ X α∣0⟩ + β∣1⟩
1
11 4
(α∣1⟩ − β∣0⟩)∣11⟩ ZX α∣0⟩ + β∣1⟩
Teleportation analysis
⎧ ⎫
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩ ⎨ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪ ⎪
⎪
⎩ H ⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎬ α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩
⎧
⎪
⎪
⎪
+ Alice
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
+ ⎪ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
Bob ⎪
⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎩ X Z ⎭

∣π0 ⟩ ∣π1 ⟩ ∣π2 ⟩

∣π0 ⟩ = √1 (α∣00⟩∣0⟩∣γ0 ⟩ + α∣11⟩∣0⟩∣γ0 ⟩ + β∣00⟩∣1⟩∣γ1 ⟩ + β∣11⟩∣1⟩∣γ1 ⟩)

2

∣π1 ⟩ = √1 (α∣00⟩∣0⟩∣γ0 ⟩ + α∣11⟩∣0⟩∣γ0 ⟩ + β∣01⟩∣1⟩∣γ1 ⟩ + β∣10⟩∣1⟩∣γ1 ⟩)

2

∣π2 ⟩ = √1 (α∣00⟩∣+⟩∣γ0 ⟩ + α∣11⟩∣+⟩∣γ0 ⟩ + β∣01⟩∣−⟩∣γ1 ⟩ + β∣10⟩∣−⟩∣γ1 ⟩)

2

1
= 2
(α∣0⟩∣00⟩∣γ0 ⟩ + α∣0⟩∣01⟩∣γ0 ⟩ + α∣1⟩∣10⟩∣γ0 ⟩ + α∣1⟩∣11⟩∣γ0 ⟩
+β∣1⟩∣00⟩∣γ1 ⟩ − β∣1⟩∣01⟩∣γ1 ⟩ + β∣0⟩∣10⟩∣γ1 ⟩ − β∣0⟩∣11⟩∣γ1 ⟩)
Teleportation analysis
⎧ ⎫
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩ ⎨ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪ ⎪
⎪
⎩ H ⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎬ α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩
⎧
⎪
⎪
⎪
+ Alice
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
+ ⎪ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
Bob ⎪
⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎪
⎪ ⎪
⎪
⎩ X Z ⎭

∣π0 ⟩ ∣π1 ⟩ ∣π2 ⟩

1
∣π2 ⟩ = 2
(α∣0⟩∣00⟩∣γ0 ⟩ + α∣0⟩∣01⟩∣γ0 ⟩ + α∣1⟩∣10⟩∣γ0 ⟩ + α∣1⟩∣11⟩∣γ0 ⟩
+β∣1⟩∣00⟩∣γ1 ⟩ − β∣1⟩∣01⟩∣γ1 ⟩ + β∣0⟩∣10⟩∣γ1 ⟩ − β∣0⟩∣11⟩∣γ1 ⟩)

ab Probability Conditional state of (B, R, A, Q) Operation on B Final state of (B, R)

1
00 4
(α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩)∣00⟩ 1 α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩
1
01 4
(α∣0⟩∣γ0 ⟩ − β∣1⟩∣γ1 ⟩)∣01⟩ Z α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩
1
10 4
(α∣1⟩∣γ0 ⟩ + β∣0⟩∣γ1 ⟩)∣10⟩ X α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩
1
11 4
(α∣1⟩∣γ0 ⟩ − β∣0⟩∣γ1 ⟩)∣11⟩ ZX α∣0⟩∣γ0 ⟩ + β∣1⟩∣γ1 ⟩
Remarks on teleportation
• Teleportation is not an application of quantum information — it’s a way to
perform quantum communication.
• Teleportation motivates entanglement distillation as a means to reliable
quantum communication.
• Beyond its potential for communication, teleportation also has fundamental
importance in the study of quantum information and computation.
2. Superdense coding
Superdense coding set-up
Scenario

Alice has two classical bits that she wishes to transmit to Bob.
• Alice is able to send a single qubit to Bob.
• Alice and Bob share an e-bit.

Remark

Without the e-bit, Alice and Bob’s task would be impossible…

Holevo’s theorem implies that two classical bits of communication cannot

be reliably transmitted by a single qubit alone.
Superdense coding protocol
Circuit description

b
a

⎧
⎪ Z X
⎪
⎪
⎪
⎪
⎪
⎪ Alice
⎪
⎪
⎪
⎪
⎪
⎪
Bob
+ ⎪
∣ϕ ⟩ ⎪
⎨
⎪
⎪
⎪
⎪
⎪ H b
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎩ + a
Superdense coding analysis
b
+ 1 1
a ∣ϕ ⟩ = √ ∣00⟩ + √ ∣11⟩
2 2
⎧
⎪ Z X 1 1
⎪
⎪ −
⎪
⎪
⎪
∣ϕ ⟩ = √ ∣00⟩ − √ ∣11⟩
⎪
⎪ Alice 2 2
⎪
⎪
⎪
⎪ Bob
⎪
+ ⎪ 1 1
∣ϕ ⟩ ⎪
⎨
⎪
+
∣ψ ⟩ = √ ∣01⟩ + √ ∣10⟩
⎪
⎪
⎪
⎪
⎪
H b 2 2
⎪
⎪
⎪
⎪
⎪
⎪
⎪ − 1 1
⎪ ∣ψ ⟩ = √ ∣01⟩ − √ ∣10⟩
⎪
⎩ + a
2 2

ab Alice’s action Bob’s action

+ + +
00 ∣ϕ ⟩ ↦ ∣ϕ ⟩ ∣ϕ ⟩ ↦ ∣00⟩
+ − −
01 ∣ϕ ⟩ ↦ ∣ϕ ⟩ ∣ϕ ⟩ ↦ ∣01⟩
+ + +
10 ∣ϕ ⟩ ↦ ∣ψ ⟩ ∣ψ ⟩ ↦ ∣10⟩
+ − −
11 ∣ϕ ⟩ ↦ ∣ψ ⟩ ∣ψ ⟩ ↦ −∣11⟩
Remarks on superdense coding
• Superdense coding seems unlikely to be useful in a practical sense.

• The underlying idea is fundamentally important, and illustrates an

interesting aspect of entanglement.

• Together with teleportation, superdense coding establishes an equivalence:

1 qubit of quantum 1 ebit 2 bits of classical

⟺
communication communication
3. The CHSH game
Nonlocal games
Mathematical abstractions of games are both important and useful.
The CHSH game is an example of a nonlocal game.

Set-up
No communication
The players are Alice and Bob, who between Alice and Bob
cooperate as a team.

The game is run by a referee.

Alice Bob
Alice and Bob can prepare for the
game however they choose… a b
x y
…but once the game starts they are
forbidden from communicating. Referee
Nonlocal games
Mathematical abstractions of games are both important and useful.
The CHSH game is an example of a nonlocal game.

The referee
No communication
The referee uses randomness to select between Alice and Bob
the questions x and y.

The referee determines whether a pair

of answers (a, b) wins or loses for the Alice Bob
questions pair (x, y) according to
some fixed rule. a b
x y
(A precise description of the referee
Referee
defines an instance of a nonlocal
game.)
Nonlocal games
CHSH game referee
1. The questions and answers are all bits:

Alice Bob
x, y, a, b ∈ {0, 1}

2. The questions x and y are chosen

a b
uniformly at random. x y
3. A pair of answers (a, b) wins for (x, y) if Referee
a⊕b=x∧y

and loses otherwise.

Deterministic strategies
No deterministic strategy can win every time.

(x, y) winning condition

(0, 0) a=b a(0) ⊕ b(0) = 0
a(0) ⊕ b(1) = 0
(0, 1) a=b
a(1) ⊕ b(0) = 0
(1, 0) a=b
a(1) ⊕ b(1) = 1
(1, 1) a=
/b
It follows that no deterministic strategy can
with with probability greater than 3/4.
Nonlocal games
CHSH game referee
1. The questions and answers are all bits:

Alice Bob
x, y, a, b ∈ {0, 1}

2. The questions x and y are chosen

a b
uniformly at random. x y
3. A pair of answers (a, b) wins for (x, y) if Referee
a⊕b=x∧y

and loses otherwise.

Probabilistic strategies
Every probabilistic strategy can be viewed as
a random choice of a deterministic strategy.
(x, y) winning condition
(0, 0) a=b It follows that no probabilistic strategy can
win with probability greater than 3/4.
(0, 1) a=b
(1, 0) a=b
(1, 1) a=
/b
Nonlocal games
CHSH game referee
∣ψ⟩
1. The questions and answers are all bits:

x, y, a, b ∈ {0, 1}

2. The questions x and y are chosen Alice Bob

uniformly at random.

3. A pair of answers (a, b) wins for (x, y) if a b

x y
a⊕b=x∧y
Referee
and loses otherwise.

(x, y) winning condition

Can a quantum strategy do better?
(0, 0) a=b
(0, 1) a=b
(1, 0) a=b
(1, 1) a=
/b
CHSH game strategy
For each angle θ (measured in radians), define a unit vector

∣ψθ ⟩ = cos(θ)∣0⟩ + sin(θ)∣1⟩

∣1⟩ √ √ √ √
2− 2 2+ 2
∣ψ3π/8 ⟩ = ∣0⟩ = ∣1⟩
2 2 θ cos(θ) sin(θ)
∣ψπ/4 ⟩ = √1 ∣0⟩ + √1 ∣1⟩
2 2
0 1 0
√ √ √ √
π 2+ 2 2− 2
8 2 2
−∣0⟩ ∣0⟩
π √1 √1
4 2 2
√ √ √ √
3π 2− 2 2+ 2
8 2 2
π
2
0 1
−∣1⟩
CHSH game strategy
For each angle θ (measured in radians), define a unit vector

∣ψθ ⟩ = cos(θ)∣0⟩ + sin(θ)∣1⟩

∣1⟩
∣ψα ⟩ θ cos(θ) sin(θ)
∣ψβ ⟩ 0 1 0
α−β
√ √ √ √
π 2+ 2 2− 2
8 2 2
−∣0⟩ ∣0⟩
π √1 √1
4 2 2
√ √ √ √
3π 2− 2 2+ 2
8 2 2
π
2
0 1
−∣1⟩

By one of the angle addition formulas we have

⟨ψα ∣ψβ ⟩ = cos(α) cos(β) + sin(α) sin(β) = cos(α − β)

+ cos(α) cos(β) + sin(α) sin(β) cos(α − β)

⟨ψα ⊗ ψβ ∣ϕ ⟩ = √ = √
2 2
CHSH game strategy
For each angle θ (measured in radians), define a unit vector

∣ψθ ⟩ = cos(θ)∣0⟩ + sin(θ)∣1⟩

Also define a unitary matrix

Uθ = ∣0⟩⟨ψθ ∣ + ∣1⟩⟨ψθ+π/2 ∣

Alice and Bob’s strategy

Alice and Bob share an e-bit (A, B).

Alice’s actions Bob’s actions

Alice applies an operation to A as Bob applies an operation to B as

follows: follows:

U0 if x = 0 Uπ/8 if y = 0
{ {
Uπ/4 if x = 1 U−π/8 if y = 1

She then measures A and sends the He then measures B and sends the
result to the referee. result to the referee.
CHSH game strategy
Alice and Bob’s strategy

Alice and Bob share an e-bit (A, B).

Alice’s actions Bob’s actions

Alice applies an operation to A: Bob applies an operation to B:

U0 if x = 0 Uπ/8 if y = 0
{ {
Uπ/4 if x = 1 U−π/8 if y = 1

She then measures A and sends the He then measures B and sends the
result to the referee. result to the referee.

⎧
⎪ U π8 U− π8 b
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
+ ⎪
Bob
∣ϕ ⟩ ⎪
⎨
⎪
⎪
⎪ Alice
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ U0 U π4 a
⎩

x
Analysis of the strategy
+ cos(α − β)
Uθ = ∣0⟩⟨ψθ ∣ + ∣1⟩⟨ψθ+π/2 ∣ ⟨ψα ⊗ ψβ ∣ϕ ⟩ = √
2

Case 1: (x, y) = (0, 0)

Alice performs U0 and Bob performs U π .

8

+ + +
(U0 ⊗ U π8 )∣ϕ ⟩ = ∣00⟩⟨ψ0 ⊗ ψ π8 ∣ϕ ⟩ + ∣01⟩⟨ψ0 ⊗ ψ 5π ∣ϕ ⟩
8
+ +
+ ∣10⟩⟨ψ π2 ⊗ ψ π8 ∣ϕ ⟩ + ∣11⟩⟨ψ π2 ⊗ ψ 5π ∣ϕ ⟩
8

cos(− π8 )∣00⟩ + cos(− 5π

8
)∣01⟩ + cos( 3π
8
)∣10⟩ + cos(− π8 )∣11⟩
= √
2

√
(a, b) Probability Simplified 2+ 2
√ Pr(a = b) = ≈ 0.85
1 2 2+ 2 4
(0, 0) 2
cos (− π8 ) 8
√
√ 2− 2
(0, 1) 1
cos
2
(− 5π ) 2− 2 Pr(a =
/ b) = ≈ 0.15
2 8 8 4
√
1 2 3π 2− 2
(1, 0) 2
cos ( 8 ) 8 √
√ 2+ 2
1 2 2+ 2 They win with probability ≈ 0.85.
(1, 1) 2
cos (− π8 ) 8
4
Analysis of the strategy
+ cos(α − β)
Uθ = ∣0⟩⟨ψθ ∣ + ∣1⟩⟨ψθ+π/2 ∣ ⟨ψα ⊗ ψβ ∣ϕ ⟩ = √
2

Case 2: (x, y) = (0, 1)

Alice performs U0 and Bob performs U− π .

8

+ + +
(U0 ⊗ U− π8 )∣ϕ ⟩ = ∣00⟩⟨ψ0 ⊗ ψ− π8 ∣ϕ ⟩ + ∣01⟩⟨ψ0 ⊗ ψ 3π ∣ϕ ⟩
8
+ +
+ ∣10⟩⟨ψ π2 ⊗ ψ− π8 ∣ϕ ⟩ + ∣11⟩⟨ψ π2 ⊗ ψ 3π ∣ϕ ⟩
8

cos( π8 )∣00⟩ + cos(− 3π

8
)∣01⟩ + cos( 5π
8
)∣10⟩ + cos( π8 )∣11⟩
= √
2

√
(a, b) Probability Simplified 2+ 2
√ Pr(a = b) = ≈ 0.85
1 2 2+ 2 4
(0, 0) 2
cos ( π8 ) 8
√
√ 2− 2
(0, 1) 1 2
cos (− 3π ) 2− 2 Pr(a =
/ b) = ≈ 0.15
2 8 8 4
√
1 2
(1, 0) 2
cos ( 5π
8
) 2− 2
8 √
√ 2+ 2
1 2 2+ 2 They win with probability ≈ 0.85.
(1, 1) 2
cos ( π8 ) 8
4
Analysis of the strategy
+ cos(α − β)
Uθ = ∣0⟩⟨ψθ ∣ + ∣1⟩⟨ψθ+π/2 ∣ ⟨ψα ⊗ ψβ ∣ϕ ⟩ = √
2

Case 3: (x, y) = (1, 0)

Alice performs U π and Bob performs U π .

4 8

+ + +
(U π4 ⊗ U π8 )∣ϕ ⟩ = ∣00⟩⟨ψ π4 ⊗ ψ π8 ∣ϕ ⟩ + ∣01⟩⟨ψ π4 ⊗ ψ 5π ∣ϕ ⟩
8
+ +
+ ∣10⟩⟨ψ 3π ⊗ ψ π8 ∣ϕ ⟩ + ∣11⟩⟨ψ 3π ⊗ ψ 5π ∣ϕ ⟩
4 4 8

cos( π8 )∣00⟩ + cos(− 3π

8
)∣01⟩ + cos( 5π
8
)∣10⟩ + cos( π8 )∣11⟩
= √
2

√
(a, b) Probability Simplified 2+ 2
√ Pr(a = b) = ≈ 0.85
1 2 2+ 2 4
(0, 0) 2
cos ( π8 ) 8
√
√ 2− 2
(0, 1) 1
cos
2
(− 3π
) 2− 2 Pr(a =
/ b) = ≈ 0.15
2 8 8 4
√
1 2 5π 2− 2
(1, 0) 2
cos ( 8 ) 8 √
√ 2+ 2
1 2 2+ 2 They win with probability ≈ 0.85.
(1, 1) 2
cos ( π8 ) 8
4
Analysis of the strategy
+ cos(α − β)
Uθ = ∣0⟩⟨ψθ ∣ + ∣1⟩⟨ψθ+π/2 ∣ ⟨ψα ⊗ ψβ ∣ϕ ⟩ = √
2

Case 4: (x, y) = (1, 1)

Alice performs U π and Bob performs U− π .

4 8

+ + +
(U π4 ⊗ U− π8 )∣ϕ ⟩ = ∣00⟩⟨ψ π4 ⊗ ψ− π8 ∣ϕ ⟩ + ∣01⟩⟨ψ π4 ⊗ ψ 3π ∣ϕ ⟩
8
+ +
+ ∣10⟩⟨ψ 3π ⊗ ψ− π8 ∣ϕ ⟩ + ∣11⟩⟨ψ 3π ⊗ ψ 3π ∣ϕ ⟩
4 4 8

cos( 3π
8
)∣00⟩ + cos(− π8 )∣01⟩ + cos( 7π
8
)∣10⟩ + cos( 3π
8
)∣11⟩
= √
2

√
(a, b) Probability Simplified 2− 2
√ Pr(a = b) = ≈ 0.15
1 2 4
(0, 0) 2
cos ( 3π
8
) 2− 2
8
√
√ 2+ 2
(0, 1) 1 2 π
cos (− 8 ) 2+ 2 Pr(a =
/ b) = ≈ 0.85
2 8 4
√
1 2
(1, 0) 2
cos ( 7π
8
) 2+ 2
8 √
√ 2+ 2
1 2 They win with probability ≈ 0.85.
(1, 1) 2
cos ( 3π
8
) 2− 2
8
4
Analysis of the strategy
y

⎧
⎪ U π8 U− π8 b
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
+ ⎪
Bob
∣ϕ ⟩ ⎪
⎨
⎪
⎪
⎪ Alice
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ U0 U π4 a
⎩

√
2+ 2
The strategy wins with probability 4
≈ 0.85 (in all four cases, and therefore overall).
Analysis of the strategy
We can also think about the strategy geometrically.

Using the formula

+ 1
⟨ψα ⊗ ψβ ∣ϕ ⟩ = √ ⟨ψα ∣ψβ ⟩
2

the probabilities of different measurement outcomes can be expressed as follows.

(x, y) = (0, 0) ∣ψ 5π ⟩ ∣ψ π2 ⟩
8

(a, b) Probability
1 2
(0, 0) 2
∣⟨ψ0 ∣ψ π8 ⟩∣ ∣ψ π8 ⟩
1 2
(0, 1) 2
∣⟨ψ0 ∣ψ 5π ⟩∣ ∣ψ0 ⟩
8
1 2
(1, 0) 2
∣⟨ψ π2 ∣ψ π8 ⟩∣
1 2
(1, 1) 2
∣⟨ψ 2 ∣ψ 5π ⟩∣
π
8
Analysis of the strategy
We can also think about the strategy geometrically.

Using the formula

+ 1
⟨ψα ⊗ ψβ ∣ϕ ⟩ = √ ⟨ψα ∣ψβ ⟩
2

the probabilities of different measurement outcomes can be expressed as follows.

(x, y) = (0, 1) ∣ψ π2 ⟩ ∣ψ 3π ⟩
8

(a, b) Probability
1 2
(0, 0) 2
∣⟨ψ0 ∣ψ− π8 ⟩∣
1 2
(0, 1) 2
∣⟨ψ0 ∣ψ 3π ⟩∣ ∣ψ0 ⟩
8
1 2
(1, 0) ∣⟨ψ π2 ∣ψ− π8 ⟩∣
2 ∣ψ− π8 ⟩
1 2
(1, 1) 2
∣⟨ψ π2 ∣ψ 3π ⟩∣
8
Analysis of the strategy
We can also think about the strategy geometrically.

Using the formula

+ 1
⟨ψα ⊗ ψβ ∣ϕ ⟩ = √ ⟨ψα ∣ψβ ⟩
2

the probabilities of different measurement outcomes can be expressed as follows.

(x, y) = (1, 0) ∣ψ 5π ⟩
8
∣ψ 3π ⟩ ∣ψ π4 ⟩
(a, b) Probability 4

(0, 0) 1
∣⟨ψ π4 ∣ψ π8 ⟩∣
2 ∣ψ π8 ⟩
2
1 2
(0, 1) 2
∣⟨ψ π4 ∣ψ 5π ⟩∣
8
1 2
(1, 0) 2
∣⟨ψ 3π ∣ψ 8 ⟩∣
π
4
1 2
(1, 1) 2
∣⟨ψ 3π ∣ψ 5π ⟩∣
4 8
Analysis of the strategy
We can also think about the strategy geometrically.

Using the formula

+ 1
⟨ψα ⊗ ψβ ∣ϕ ⟩ = √ ⟨ψα ∣ψβ ⟩
2

the probabilities of different measurement outcomes can be expressed as follows.

(x, y) = (1, 1) ∣ψ 3π ⟩
8
∣ψ 3π ⟩ ∣ψ π4 ⟩
(a, b) Probability 4

1 2
(0, 0) 2
∣⟨ψ π4 ∣ψ− π8 ⟩∣
1 2
(0, 1) 2
∣⟨ψ π4 ∣ψ 3π ⟩∣
8
1 2
(1, 0) 2
∣⟨ψ 3π ∣ψ ⟩∣
−π
∣ψ− π8 ⟩
4 8
1 2
(1, 1) 2
∣⟨ψ 3π ∣ψ 3π ⟩∣
4 8
Remarks on the CHSH game
• The CHSH game is not always described as a game — it’s often described as
an experiment, or an example of a Bell test.
• The CHSH game offers a way to experimentally test the theory of quantum
information.
(The 2022 Nobel Prize in Physics was awarded to Alain Aspect, John Clauser,
and Anton Zeilinger for experiments that do this with entangled photons.)
• The study of nonlocal games more generally is a fascinating and active area
of research that still holds many mysteries.