Quantum Optimization of Currency Arbitrage via Graph-Informed Entanglement Strategies

Detecting arbitrage opportunities in the foreign exchange market is computationally challenging, as the problem’s NP-completeness causes complexity to grow rapidly with the size of the currency network. While previous studies often relied on simplified models that overlook real-world trading frictions & microstructures, this paper explicitly incorporates the real-world elements: bid and ask rates, into the QUBO formulation and evaluates performance under QAOA with various entanglement strategies. Results show that a modular, graph-informed entanglement structure achieves near-optimal approximation ratios (approximately 0.999). By integrating market spread dynamics into the quantum optimization process, this work advances beyond existing arbitrage models and demonstrates the broader applicability of graph-based quantum algorithms for financial optimization.