Variational Quantum Eigensolver for Real-World Finance: Scalable Solutions for Dynamic Portfolio Optimization Problems

Battle Count: 76

Directly addresses dynamic portfolio optimization with realistic financial constraints (transaction costs, risk aversion, investment bounds, MARR thresholds). Uses real market data (IBEX 35). Demonstrates competitive financial performance vs CPLEX. However, practical deployment is limited by current quantum hardware availability and the methodology is still in research/proof-of-concept stage rather than production-ready.

Requires access to IBM Quantum hardware (IBM Fez QPU with 156 qubits), Qiskit Runtime with specific primitives (estimator, sampler), custom ansatz design matching QPU topology, DE optimizer configuration, ISQR post-processing pipeline with probabilistic bit-flipping, VQEC sequential decomposition, and normalization procedures. The full pipeline involves quantum-classical hybrid workflows with multiple speedup optimizations. Not feasible without quantum hardware access.

The paper provides detailed formulation, ansatz design, DE parameters, ISQR configuration (Ns=10^6, M=10 batches, Nb=10^5), asset parameters (Table VII), and uses publicly accessible IBM Fez QPU. However, quantum hardware access is limited and results may vary with hardware noise. No code repository is mentioned. Classical benchmarks (CPLEX, PyPortfolioOpt) are reproducible.