Source code for qiskit_nature.second_q.mappers.bosonic_logarithmic_mapper

# This code is part of a Qiskit project.
#
# (C) Copyright IBM 2024, 2025.
#
# This code is licensed under the Apache License, Version 2.0. You may
# obtain a copy of this license in the LICENSE.txt file in the root directory
# of this source tree or at http://www.apache.org/licenses/LICENSE-2.0.
#
# Any modifications or derivative works of this code must retain this
# copyright notice, and modified files need to carry a notice indicating
# that they have been altered from the originals.

"""The Logarithmic Mapper for Bosons."""

from __future__ import annotations
import operator
import math
import logging

from functools import reduce, lru_cache

import numpy as np

from qiskit.quantum_info import SparsePauliOp

from qiskit_nature.second_q.operators import BosonicOp
from .bosonic_mapper import BosonicMapper

logger = logging.getLogger(__name__)


[docs] class BosonicLogarithmicMapper(BosonicMapper): """The Logarithmic boson-to-qubit Mapper. This mapper generates a logarithmic encoding of the Bosonic operator :math:`b_k^\\dagger, b_k` to qubit operators (linear combinations of pauli strings). In this logarithmic encoding the number of qubits necessary to represent a bosonic mode is determined by the max occupation :math:`n_k^{max}` of the mode (meaning the number of states used in the expansion of the mode, or equivalently the state at which the maximum excitation can take place). The number of qubits is given by: :math:`\\lceil\\log_2(n_k^{max} + 1)\\rceil`. .. note:: A consequence of the rounding up for determining the number of required qubits is that the actual max occupation is often larger than the one selected by the user. For example, if the user selects :math:`n_k^{max} = 2`, then the number of required qubits is :math:`\\lceil\\log_2(3)\\rceil = 2`. If we now compute the max occupation for 2 qubits, we get :math:`2^2 - 1 = 3`, which is larger than the user-selected max occupation. The user should expect that the actual max occupation is always larger than or equal to the one selected. If the code changes the max occupation, a warning will appear in the logs. The mode :math:`|k\\rangle` is then mapped to the occupation number vector :math:`|0_{n_k^{max}}, 0_{n_k^{max} - 1},..., 0_{n_k + 1}, 1_{n_k}, 0_{n_k - 1},..., 0_{0_k}\\rangle` This class implements the equation (34) and (35) of Reference [1]. .. math:: b_k^\\dagger = \\sum_{n_k = 0}^{2^{N_q}-2}\\left(\\sqrt{n_k + 1}|n+1\\rangle\\langle n|\\right) b_k = \\sum_{n_k = 1}^{2^{N_q}-1}\\left(\\sqrt{n_k}|n-1\\rangle\\langle n|\\right) where :math:`N_q` is the number of qubits used to represent each mode (given by :math:`\\lceil\\log_2(n_k^{max} + 1)\\rceil`). This implementation first computes each :math:`|n+1\\rangle\\langle n|` and :math:`|n-1\\rangle\\langle n|` in a binary representation and then uses equation (37) from Reference [1] to map to the Pauli operators. The length of the qubit register is: .. code-block:: python BosonicOp.num_modes * math.ceil(numpy.log2(BosonicLogarithmicMapper.max_occupation + 1)) Below is an example of how one can use this mapper: .. code-block:: python from qiskit_nature.second_q.mappers import BosonicLogarithmicMapper from qiskit_nature.second_q.operators import BosonicOp mapper = BosonicLogarithmicMapper(max_occupation=2) qubit_op = mapper.map(BosonicOp({'+_0 -_0': 1}, num_modes=1)) .. note:: Since this mapper truncates the maximum occupation of a bosonic state as represented in the qubit register, the commutation relations after the mapping differ from the standard ones. Please refer to Section 4, equation 22 of Reference [2] for more details. References: [1] Bo Peng et al., Quantum Simulation of Boson-Related Hamiltonians: Techniques, Effective Hamiltonian Construction, and Error Analysis, Arxiv https://doi.org/10.48550/arXiv.2307.06580 [2] R. Somma et al., Quantum Simulations of Physics Problems, Arxiv https://doi.org/10.48550/arXiv.quant-ph/0304063 """ def __init__(self, max_occupation: int) -> None: # Compute the actual max occupation from the one selected by the user self.number_of_qubits_per_mode: int = ( 1 if max_occupation == 0 else math.ceil(np.log2(max_occupation + 1)) ) max_calculated_occupation = 2**self.number_of_qubits_per_mode - 1 if max_occupation != max_calculated_occupation: # Use of a message variable to suppress pylint warning logging-not-lazy warning_message = ( f"The user requested a max occupation of {max_occupation}, " + f"but the actual max occupation is {max_calculated_occupation}." ) logger.warning(warning_message) super().__init__(max_calculated_occupation) @property def number_of_qubits_per_mode(self) -> int: """The minimum number of qubits required to represent a bosonic mode given a max occupation.""" return self._number_of_qubits_per_mode @number_of_qubits_per_mode.setter def number_of_qubits_per_mode(self, num_qubits: int) -> None: if num_qubits < 1: raise ValueError(f"The number of qubits must be at least 1, and not {num_qubits}.") self._number_of_qubits_per_mode: int = num_qubits def _map_single( self, second_q_op: BosonicOp, *, register_length: int | None = None ) -> SparsePauliOp: """Maps a :class:`~qiskit_nature.second_q.operators.SparseLabelOp` to a ``SparsePauliOp``. Args: second_q_op: the ``SparseLabelOp`` to be mapped. register_length: when provided, this will be used to overwrite the ``register_length`` attribute of the operator being mapped. This is possible because the ``register_length`` is considered a lower bound in a ``SparseLabelOp``. Returns: The qubit operator corresponding to the problem-Hamiltonian in the qubit space. Raises: ValueError: if any term in the bosonic operator is not in the form `+_k` or `-_k`. """ if register_length is None: register_length = second_q_op.num_modes # The actual register length is the number of qubits per mode times the number of modes qubit_register_length: int = register_length * self.number_of_qubits_per_mode # Create a Pauli operator, which we will fill in this method pauli_op: list[SparsePauliOp] = [] # Then we loop over all the terms of the bosonic operator for terms, coeff in second_q_op.terms(): # Then loop over each term (terms -> List[Tuple[string, int]]) bos_op_to_pauli_op = SparsePauliOp(["I" * qubit_register_length], coeffs=[1.0]) # Loop over the operators in the term for op, idx in terms: if op not in ("+", "-"): raise ValueError( f"Invalid bosonic operator: `{op}_{idx}`." "All bosonic operators must have the following shape: `+_k` or `-_k`." ) pauli_expansion: list[SparsePauliOp] = [] # Define the index of the mode in the qubit register mode_index_in_register: int = idx * self.number_of_qubits_per_mode # Now we start mapping the operator. First, define the range of the sum terms_range = ( range(2**self.number_of_qubits_per_mode - 1) if op == "+" else range(1, 2**self.number_of_qubits_per_mode) ) for n in terms_range: # In each iteration we deal with a term of the form sqrt(n+1)*|n+1><n| or # sqrt(n)*|n-1><n|. The initial and final states are represented in binary. # Define the prefactor and the initial and final states (which results from the # action of the operator). They vary depending on the operator prefactor = np.sqrt(n + 1) if op == "+" else np.sqrt(n) final_state: str = ( f"{(n + 1):0{self.number_of_qubits_per_mode}b}" if op == "+" else f"{(n - 1):0{self.number_of_qubits_per_mode}b}" ) init_state: str = f"{n:0{self.number_of_qubits_per_mode}b}" # Now build the Pauli operators single_mapped_term = SparsePauliOp(["I" * qubit_register_length], coeffs=[1.0]) # pylint: disable=consider-using-enumerate for j in range(len(init_state)): # We need to comply to the little endian notation of qiskit. # For the binary string representation of the state, the first element is the # most significant bit. Thus, it needs to be put to the end of the mode in the # qubit register. i: int = len(init_state) - j - 1 # Get the Pauli operator for the single qubit of the term we are mapping and # compose it with the already mapped ones single_mapped_term = single_mapped_term.compose( self._get_single_qubit_pauli_matrix( mode_index_in_register + i, qubit_register_length, final_state[j] + init_state[j], ) ) pauli_expansion.append(prefactor * single_mapped_term) # Add the Pauli expansion for a single n_k to map of the bosonic operator bos_op_to_pauli_op = reduce(operator.add, pauli_expansion).compose( bos_op_to_pauli_op ) # Add the map of the single boson op (e.g. +_0) to the map of the full bosonic operator pauli_op.append(coeff * reduce(operator.add, bos_op_to_pauli_op.simplify())) # return the lookup table for the transformed XYZI operators return reduce(operator.add, pauli_op) @lru_cache(maxsize=32) def _get_single_qubit_pauli_matrix( self, qubit_idx: int, register_length: int, qubit_operator: str ) -> SparsePauliOp: """This method builds the Qiskit Pauli operators of one of the operators: I_+ = I + Z, I_- = I - Z, S_+ = X + iY and S_- = X - iY. Args: qubit_idx: the register index of the qubit on which the operator is acting. register_length: the length of the qubit register.\n qubit_operator: the operator to be mapped. Possible values are: - '00', which corresponds to '|0><0|' or 'I+' - '11', which corresponds to '|1><1|' or 'I-' - '01', which corresponds to '|0><1|' or 'S+' - '10', which corresponds to '|1><0|' or 'S-' Returns: A SparsePauliOp representing the Pauli operator. """ if qubit_operator == "00": # I+ return SparsePauliOp.from_sparse_list( [("", [], 0.5), ("Z", [qubit_idx], 0.5)], num_qubits=register_length ) if qubit_operator == "11": # I- return SparsePauliOp.from_sparse_list( [("", [], 0.5), ("Z", [qubit_idx], -0.5)], num_qubits=register_length ) if qubit_operator == "01": # S+ return SparsePauliOp.from_sparse_list( [("X", [qubit_idx], 0.5), ("Y", [qubit_idx], 0.5j)], num_qubits=register_length ) if qubit_operator == "10": # S- return SparsePauliOp.from_sparse_list( [("X", [qubit_idx], 0.5), ("Y", [qubit_idx], -0.5j)], num_qubits=register_length ) raise ValueError( f"Invalid operator {qubit_operator}. Possible values are '00', '11', '01' and '10'." )