Quantum computing, as an emerging computational paradigm, has garnered significant attention due to its potential advantages in combinatorial optimization and quantum chemistry problems A hybrid quantum-classical model that integrates VQE and QAOA for efficient task assignment and path optimization is suggested. However, constrained by hardware noise, qubit limitations, and parameter optimization bottlenecks, these algorithms still face challenges such as convergence instability, gradient vanishing, and low ground state probability in large-scale problems. This paper proposes a hybrid optimization framework that integrates improved VQE and QAOA, introducing adaptive parameter initialization and multi-level parameter scheduling mechanisms into VQE. In contrast, QAOA combines an improved classical optimizer and parameter perturbation strategy, thereby enhancing the algorithm’s global convergence capability in the search space. Experiments on combined optimization examples with 6, 9, 10, and 12 qubits demonstrate that this method can significantly improve ground state probability and convergence stability under limited hardware conditions. The results validate the effectiveness of the proposed framework in complex scheduling and resource allocation tasks, showcasing its potential applications in the practical implementation of quantum optimization.
Xu et al. (Fri,) studied this question.