Quantum scrambling and teleportation

08 Jun 2019

$\newcommand{\ket}{\left|{#1}\right\rangle}$ $\newcommand\bra[1]{\langle{#1}}$ $\newcommand{\norm}[1]{\left\lVert#1\right\rVert}$

Quantum information scrambling is when local information is thrown into a disorder throughout a quantum system. This phenomena is similar to quantum decoherence where the information in a system is lost to the environment. Another way of viewing this incident is the qubits of the system entangle; and thus correlate to the qubits in the environment. The problem is that quantum dechorence happens naturally to quantum computers since qubits can’t be perfectly isolated from the environment. This makes an experiment that tests for quantum information scrambling difficult because it seems like there is no way to distinguish if a system dispersed it’s information throughout itself or the environment without resorting to quantum state tomography; which is intractable as the size of the quantum system grows. Fortunately, there is another way to differentiate between a quantum scrambled system and a quantum decohered one by means of quantum teleportation.

Quantum teleportation is the problem of Alice needing to transport her state to Bob using classical communication and an EPR pair shared by both parties. Unlike the standard quantum teleportation protocol where the fidelity of sending the unknown quantum state is 100%. The protocol talked about here is probabilistic. First, we define the many-body-system by creating 3 EPR pairs, then we apply $U$ to half of the system and $U^\dagger$ to the other half where $U$ (a unitary that delocalizes all single qubit operators into three-qubit operators) performs the scrambling. Since Alice's quantum state is fully scrambled in the system, Bob can sucessfully decode it by making a Bell measurement of any pair of qubits.

The quantum circuit and program that defines the many-body-system, Alice's state, the scrambling of Alice's state and it's decoding by Bob is displayed below. The success of this prototcol is usually around 50%.

0: ---X----------@----------@---H-------@---@-------------------
                 |          |           |   |
1: ---H-----@----|----------@---@---H---|---@---@---@---H---M---
            |    |              |       |       |   |
2: ---H----@|----@--------------@---H---@-------@---X---M-------
           ||
3: --------X|----@--------------@---H---@-------@---------------
            |    |              |       |       |
4: ---------X----|----------@---@---H---|---@---@---------------
                 |          |           |   |
5: ---H----@-----@----------@---H-------@---@-------------------
           |
6: --------X-----M('bob')---------------------------------------
"""To demonstrate quantum scrambling using teleportation without Grover search"""
import numpy as np
import cirq

def bell_basis_measurement(first_qubit, second_qubit):
    circuit = cirq.Circuit()
    circuit.append([
        cirq.CNOT(first_qubit, second_qubit),
        cirq.H(first_qubit),
        cirq.measure(first_qubit), 
        cirq.measure(second_qubit),   
    ])
    return circuit

def scrambling_circuit(q0, q1, q2, q3, q4, q5):
    circuit = cirq.Circuit()
    
    circuit.append([
        # Define U
        cirq.CZ(q0, q2),
        cirq.CZ(q0, q1),
        cirq.CZ(q1, q2),
        cirq.H(q0),
        cirq.H(q1),
        cirq.H(q2),
        cirq.CZ(q0, q2),
        cirq.CZ(q0, q1),
        cirq.CZ(q1, q2),

        # Define conjugate of U
        cirq.CZ(q3, q5),
        cirq.CZ(q4, q5),
        cirq.CZ(q3, q4),
        cirq.H(q3),
        cirq.H(q4),
        cirq.H(q5),
        cirq.CZ(q3, q5),
        cirq.CZ(q4, q5),
        cirq.CZ(q3, q4),
    ])

    return circuit


def scrambling_teleportation(circuit, qubits_to_measure):
    """
    Output exmple: Fidelity 48/100
    """
    repetitions = 100

    # Prepare the state to be |1>   
    circuit.append([
        cirq.X(q0),
    ])

    # Defining many-body-system
    circuit.append([
        cirq.H(q1),
        cirq.H(q2),
        cirq.H(q5),
        cirq.CNOT(q5, q6),
        cirq.CNOT(q2, q3),
        cirq.CNOT(q1, q4),    
    ])

    # Retrieving U and and conjugate of U on the subsystems
    scrambling_unitary = scrambling_circuit(q0, q1, q2, q3, q4, q5)

    circuit.append([
        scrambling_unitary,
        bell_basis_measurement(qubits_to_measure[0], qubits_to_measure[1]), 
        cirq.measure(q6, key='bob')
    ])
   
    results = cirq.Simulator().run(circuit, repetitions=repetitions)
    bob_measurement = np.array(results.measurements['bob'][:, 0])
    print("Fidelity {}/{}".format(sum(bob_measurement), repetitions))
    
   


if __name__ == '__main__':
    q0, q1, q2, q3, q4, q5, q6 = cirq.LineQubit.range(7)
    circuit = cirq.Circuit()
    # Any Bell Pair measurement should give the same fidelities (aound 50%)
    scrambling_teleportation(circuit, qubits_to_measure=[q1, q2])
    

Discuss on Github

Reference

To the extent possible under law, Victory Omole has waived all copyright and related or neighboring rights to this work by licensing it under a Public Domain License.