Source code for mqt.bench.benchmarks.qaoa

# Copyright (c) 2023 - 2026 Chair for Design Automation, TUM
# Copyright (c) 2025 - 2026 Munich Quantum Software Company GmbH
# All rights reserved.
#
# SPDX-License-Identifier: MIT
#
# Licensed under the MIT License

"""QAOA benchmark definition. Code is based on https://github.com/qiskit-community/qiskit-application-modules-demo-sessions/blob/main/qiskit-optimization/qiskit-optimization-demo.ipynb."""

from __future__ import annotations

import numpy as np
from qiskit.circuit import ParameterVector, QuantumCircuit

from ._registry import register_benchmark




@register_benchmark("qaoa", description="Quantum Approximation Optimization Algorithm (QAOA)")
def create_circuit(num_qubits: int, repetitions: int = 2, seed: int = 10) -> QuantumCircuit:
    """Constructs a quantum circuit implementing QAOA for a Max-Cut example with random parameters.

    Arguments:
        num_qubits: Number of qubits in the circuit (equal to the number of graph nodes).
        repetitions: Number of QAOA layers (repetitions of the ansatz).
        seed: Seed for reproducibility.

    Returns:
        QuantumCircuit: Quantum circuit implementing QAOA.
    """
    rng = np.random.default_rng(seed)
    adjacency_matrix = rng.integers(0, 2, size=(num_qubits, num_qubits))
    adjacency_matrix = np.triu(adjacency_matrix, 1)  # Upper triangular part for undirected graph

    gamma_values = ParameterVector("g", repetitions)
    beta_values = ParameterVector("b", repetitions)

    # Initialize QAOA circuit
    qc = QuantumCircuit(num_qubits)

    # Start in uniform superposition
    qc.h(range(num_qubits))

    # Define cost and mixer operators for each layer
    for layer in range(repetitions):
        # Cost Hamiltonian
        for i in range(num_qubits):
            for j in range(i + 1, num_qubits):
                if adjacency_matrix[i, j] != 0:
                    qc.rzz(2 * gamma_values[layer], i, j)

        # Mixer Hamiltonian
        for i in range(num_qubits):
            qc.rx(2 * beta_values[layer], i)

    qc.name = "qaoa"

    return qc