Quantum and quantum-inspired computations have shown remarkable potential in addressing complex optimization problems. Quantum annealing (QA), in particular, has demonstrated effectiveness in solving combinatorial optimization tasks and is gaining increasing attention. Meanwhile, quantuminspired computing (QIC) algorithms offer a promising approach to leveraging quantum principles on classical hardware. This study presents the first integrated hybrid framework that combines the strengths of both quantum and classical computing for multi-objective weighted portfolio optimization, a complex real-world application. In the proposed approach, QA employs scalarization techniques to rapidly approximate the optimal front, while QIC enhances solution diversity through an entanglement move mechanism that uncovers a broader set of non-dominated portfolio solutions. Experimental results indicate that the hybrid approach achieves higher-quality solutions and provides a greater variety of investment options for investors, compared to QA alone and classical multi-objective optimization algorithms. These findings highlight the feasibility and scalability of integrating quantum and classical techniques to effectively address real-world investment optimization challenges.
With the development of intelligent algorithms, new challenges have emerged in the financial field in optimizing the investment portfolio. In view of the current problems existing in portfolio optimization, the particle swarm algorithm (Particle Swarm Optimization, PSO) and quantum approximation optimization algorithm (Quantum Approximate Optimization Algorithm, QAOA) of the initial population Qanta coding is improved, and a hybrid algorithm suitable for solving enterprise portfolio optimization problems is proposed. The results found that the average performance of this method is below 0.06, which is better compared with the comparison algorithm. At the same time, with the increase of the risk level, the maximum, minimum and average increase of the yield are
10.99%,19.80% and 7.98% respectively, stabilizing. The results show that the research algorithm is effective in solving the enterprise portfolio optimization problem, has better optimization ability in practical application, and can provide reference for enterprise investors to make investment decisions.