import pennylane as qml
from pennylane.operation import Channel
import numpy as np
import math
eps = 1e-7 # constant
[docs]def max_entangled_state(settings, wires):
"""Ansatz function for maximally entangled two-qubit state preparation.
A general ``qml.Rot`` unitary is applied to one-side of a
Bell state creating a general parameterization of all maximally
entangled states.
:param settings: A list with three elements parameterizing a general single-qubit unitary
:type settings: list[float]
:param wires: The two wires on which the maximally entangled state is prepared.
:type wires: qml.Wires
"""
# prepare bell state
qml.Hadamard(wires=wires[0])
qml.CNOT(wires=wires[0:2])
# perform general rotation on first qubit
qml.Rot(*settings, wires=wires[0])
[docs]def nonmax_entangled_state(settings, wires):
"""Initializes a nonmaximally entangled GHZ-like state that has a bias between its two outcomes.
The state takes the form
.. math::
|\\psi_\\theta\\rangle = \\cos(\\frac{\\theta}{2})|0\\dots 0\\rangle + \\sin(\\frac{\\theta}{2})|1 \\dots 1\\rangle
:param settings: A list of length 1 containing paraameter :math:`\\theta`
:type settings: list[float]
:param wires: A list of wires to prepare the nonmaximally entangled GHZ-like state.
:type wires: list[int] or qml.Wires
"""
qml.RY(settings[0], wires=wires[0])
for i in range(1, len(wires)):
qml.CNOT(wires=[wires[0], wires[i]])
[docs]def bell_state_copies(settings, wires):
"""Initializes :math:`n` Bell states on :math:`2n` wires.
The first :math:`n` wires represent Alice's half of the entangled
state while Bob's half consists of the remaining :math:`n` wires.
Entanglement is shared between Alice and Bob, however, the independent
Bell states are not initially entangled with each other.
:param settings: A placeholder parameter that is not used.
:type settings: list[empty]
:param wires: The wires on which the bell states are prepared.
:type wires: qml.Wires
"""
for i in range(len(wires) // 2):
qml.Hadamard(wires=wires[i])
qml.CNOT(wires=[wires[i], wires[i + len(wires) // 2]])
[docs]def ghz_state(settings, wires):
"""Initializes a GHZ state on the provided wires.
:param settings: A placeholder parameter that is not used.
:type settings: list[empty]
:param wires: The wires on which the GHZ state is prepared.
:type wires: qml.Wires
"""
qml.Hadamard(wires=wires[0])
for i in range(1, len(wires)):
qml.CNOT(wires=[wires[0], wires[i]])
[docs]def W_state(settings, wires):
"""Initializes the three-qubit :math:`W`-state on the specified wires.
.. math::
|\\psi^W\\rangle = \\frac{1}{\\sqrt{3}}(|100\\rangle + |010\\rangle + |001 \\rangle)
:param settings: A placeholder parameter that is not used.
:type settings: list[empty]
:param wires: The wires on which the :math:`W`-state is prepared.
:type wires: qml.Wires
"""
phi = 2 * np.arccos(1 / np.sqrt(3))
qml.RY(phi, wires=wires[0])
qml.CRY(np.pi / 2, wires=wires[0:2])
qml.CNOT(wires=wires[1:3])
qml.CNOT(wires=wires[0:2])
qml.PauliX(wires=wires[0])
[docs]def graph_state_fn(edges):
"""Constructs a quantum function that prepares a graph state
where each qubit wire is a vertex and the ``edge`` are tuple pairs
of interacting wires. A graph state takes the form
.. math::
|\\psi^G \\rangle = \\prod_{(a,b) \\in Edges} CZ^{(a,b)} |+\\rangle^{\\otimes N}
where :math:`N` is the number of qubits, :math:`|+\\rangle = \\frac{1}{\\sqrt{2}}(|0\\rangle + |1\\rangle)`, :math:`(a,b)` is a pair of wires defining
an edge, and :math:`CZ^{(a,b)}` is a controlled-phase operation operating upon the qubit
pair :math:`(a,b)`. For more details, see the :meth:`qml.CZ` function in the `PennyLane docs <https://docs.pennylane.ai/en/stable/>`_.
:param edges: A list of qubit pair tuples defining the wires to which controlled-phase
operations are applied.
:type settings: list[tuple[int]]
:returns: A quantum circuit `graph_state(settings, wires)` that prepares the specified
graph state. Note that ``wires`` must contain all qubits specified in ``edges``.
:rtype: Function
"""
def graph_state(settings, wires):
for wire in wires:
qml.Hadamard(wire)
for edge in edges:
qml.CZ(wires=[wires[edge[0]], wires[edge[1]]])
return graph_state
[docs]def shared_coin_flip_state(settings, wires):
"""Initializes a state that mirrors a biased coin flip shared amongst the qubits
on ``wires[0:-1]`` where ``wires[-1]`` is an ansatz used to generate shared randomness.
The shared coin flip is represented by the density matrix.
.. math::
\\rho_\\theta = \\cos^2(\\frac{\\theta}{2})|0\\dots 0\\rangle\\langle 0 \\dots 0| + \\sin^2(\\frac{\\theta}{2})|1\\dots 1 \\rangle \\langle 1 \\dots 1|
:param settings: A list of 1 real value.
:type settings: list[float]
:param wires: The wires used to prepare the shaared coin flip state. Note that the last wire is used as an ancilla.
:type wires: qml.Wires
"""
nonmax_entangled_state(settings, wires)
[docs]def local_RY(settings, wires):
"""Performs a rotation about :math:`y`-axis on each qubit
specified by ``wires``.
:param settings: A list of ``len(wires)`` real values.
:type settings: list[float]
:param wires: The wires on which the rotations are applied.
:type wires: qml.Wires
"""
qml.broadcast(qml.RY, wires, "single", settings)
[docs]def local_Rot(settings, wires):
"""Performs an arbitrary qubit unitary as defined by :meth:`qml.Rot`
on each qubit specified by ``wires``.
For more details on :meth:`qml.Rot` please refer to the
`PennyLane docs <https://docs.pennylane.ai/en/stable/>`_.
:param settings: A list of ``3 * len(wires)`` real values.
:type settings: list[float]
:param wires: The wires to which the qubit unitaries are applied.
:type wires: qml.Wires
"""
for i in range(len(wires)):
qml.Rot(*settings[3 * i : 3 * i + 3], wires=wires[i])
[docs]def local_RXRY(settings, wires):
"""Performs a rotation about the :math:`x` and :math:`y` axes on
each qubit specified by ``wires``.
:param settings: A list of ``2*len(wires)`` real values.
:type settings: list[float]
:param wires: The wires on which the rotations are applied.
:type wires: qml.Wires
"""
for i, wire in enumerate(wires):
qml.RX(settings[2 * i], wires=wire)
qml.RY(settings[2 * i + 1], wires=wire)
[docs]def pure_amplitude_damping(noise_params, wires):
"""Implements a single qubit amplitude damping channel as a two-qubit gate.
This allows amplitude damping noise to be applied on quantum hardware or on a
state-vector simulator such as ``"default.qubit"``.
This method is equivalent to the
`pennylane.AmplitudeDamping <https://pennylane.readthedocs.io/en/stable/code/api/pennylane.AmplitudeDamping.html>`_
method. The corresponding Kraus operators are expressed as:
.. math::
K_0 = \\begin{pmatrix}1 & 0 \\\\ 0 & \\sqrt{1-\\gamma} \\end{pmatrix}, \\quad
K_1 = \\begin{pmatrix}0 & \\sqrt{\\gamma} \\\\ 0 & 0 \\end{pmatrix}.
:param noise_params: The parameter ``[gamma]`` quantifying the amount of amplitude damping noise.
:type noise_params: List[Float]
:param wires: Two wires on which to implement the amplitude damping channel.
The channel input and output is on ``wires[0]`` while ``wires[1]``
serves as the ancilla register.
:type wires: qml.Wires
"""
ry_setting = 2 * math.asin(math.sqrt(noise_params[0]))
qml.ctrl(qml.RY, control=wires[0])(ry_setting, wires=wires[1])
qml.CNOT(wires=[wires[1], wires[0]])
[docs]def pure_phase_damping(noise_params, wires):
"""Implements a single qubit phase damping channel as a two-qubit gate.
This allows phase damping noise to be applied on quantum hardware or on a
state-vector simulator such as ``"default.qubit"``.
This method is equivalent to the
`pennylane.PhaseDamping <https://pennylane.readthedocs.io/en/stable/code/api/pennylane.PhaseDamping.html>`_
method. The corresponding Kraus operators are expressed as:
.. math::
K_0 = \\begin{pmatrix}1 & 0 \\\\ 0 & \\sqrt{1-\\gamma} \\end{pmatrix}, \\quad
K_1 = \\begin{pmatrix}0 & 0 \\\\ 0 & \\sqrt{\\gamma} \\end{pmatrix}.
:param noise_params: The parameter ``[gamma]`` quantifying the amount of phase damping noise.
:type noise_params: List[Float]
:param wires: Two wires on which to implement the phase damping channel.
The channel input and output is on ``wires[0]`` while ``wires[1]``
serves as the ancilla register.
:type wires: qml.Wires
"""
ry_setting = 2 * math.asin(math.sqrt(noise_params[0]))
qml.ctrl(qml.RY, control=wires[0])(ry_setting, wires=wires[1])
[docs]class two_qubit_depolarizing(Channel):
"""Applies a two-qubit depolarizing channel using Kraus operators.
The channel is called as a quantum function ``two_qubit_depolarizing(gamma, wires)``
:param gamma: The amount of depolarizing noise in the channel.
:type gamma: Float
:param wires: Two wires on which to apply the depolarizing noise.
:type wires: qml.Wires
For a noise parameter :math:`\\gamma`, the two-qubit depolarizing
noise model is represented by the following function:
.. math::
\\mathcal{N}(\\rho) = (1-\\frac{16}{15}\\gamma)\\rho + (\\frac{16}{15})(\\frac{\\gamma}{4})) \\mathbb{I}
:raises ValueError: if ``0 <= gamma <= 1`` is not satisfied.
"""
num_params = 1
num_wires = 2
grad_method = "F"
@staticmethod
def compute_kraus_matrices(gamma):
"""Kraus matrices representing the ``two_qubit_depolarizing`` channel.
:param gamma: The amount of depolarizing noise in the channel.
:type gamma: Float
:returns: The Kraus matrices.
:rtype: List[Array]
"""
if not 0.0 - eps <= gamma <= 1.0 + eps:
raise ValueError("gamma must be in the interval [0,1].")
elif np.isclose(gamma, 0):
gamma = 0
paulis = [
np.array([[1, 0], [0, 1]]),
np.array([[0, 1], [1, 0]]),
np.array([[0, -1j], [1j, 0]]),
np.array([[1, 0], [0, -1]]),
]
kraus_ops = []
for i in range(4):
for j in range(4):
scalar = np.sqrt(1 - gamma) if i + j == 0 else np.sqrt(gamma / 15)
kraus_ops.append(scalar * np.kron(paulis[i], paulis[j]))
return kraus_ops
[docs]class colored_noise(Channel):
"""Applies a two-qubit colored noise channel using Kraus operators.
The channel is called as a quantum function ``colored_noise(gamma, wires)``.
:param gamma: The amount of colored noise in the channel.
:type gamma: Float
:param wires: Two wires on which to apply the colored noise.
:type wires: qml.Wires
For the noise parameter :math:`\\gamma`, the colored noise model is
represented by the following function:
.. math::
\\mathcal{N}(\\rho) = (1-\\gamma)\\rho + \\frac{\\gamma}{2}(|01\\rangle\\langle 01| + |10\\rangle\\langle 10|)
:raises ValueError: if ``0 <= gamma <= 1`` is not satisfied.
"""
num_params = 1
num_wires = 2
grad_method = "F"
@staticmethod
def compute_kraus_matrices(gamma):
"""Kraus matrices representing the ``colored_noise`` channel.
:param gamma: The amount of colored noise in the channel.
:type gamma: Float
:returns: The Kraus matrices.
:rtype: List[Array]
"""
if not 0.0 - eps <= gamma <= 1.0 + eps:
raise ValueError("gamma must be in the interval [0,1].")
elif np.isclose(gamma, 0):
gamma = 0
phi_plus = np.array([1, 0, 0, 1]) / np.sqrt(2)
phi_min = np.array([1, 0, 0, -1]) / np.sqrt(2)
psi_plus = np.array([0, 1, 1, 0]) / np.sqrt(2)
psi_min = np.array([0, 1, -1, 0]) / np.sqrt(2)
scalar = np.sqrt(gamma / 2)
K0 = np.sqrt(1 - gamma) * np.eye(4)
K1 = scalar * np.outer(psi_plus, phi_plus)
K2 = scalar * np.outer(psi_plus, phi_min)
K3 = scalar * np.outer(psi_plus, psi_plus)
K4 = scalar * np.outer(psi_plus, psi_min)
K5 = scalar * np.outer(psi_min, phi_plus)
K6 = scalar * np.outer(psi_min, phi_min)
K7 = scalar * np.outer(psi_min, psi_plus)
K8 = scalar * np.outer(psi_min, psi_min)
return [K0, K1, K2, K3, K4, K5, K6, K7, K8]