arXiv:2507.07457 · Advanced Quantum Technologies
Quantum Approximate Optimization Algorithm for a Frustrated Ising Model: Simulations and Quantum‐Hardware Experiments
Seunghan Lee, Hunpyo Lee
Abstract
ABSTRACT A frustrated Ising model on a two‐dimensional square lattice is investigated using exact diagonalization and the Quantum Approximate Optimization Algorithm (QAOA), with additional benchmarks on IonQ quantum hardware. The dependence of the ground‐state sampling probability on the number of QAOA layers and the frustration strength is analyzed. In weakly frustrated regimes, the probability increases rapidly with a modest increase in the number of QAOA layers. In contrast, in the strongly frustrated regime, the probability remains limited even at larger circuit depths, reflecting the enhanced degeneracy and low‐energy competition induced by frustration. This behavior indicates that strongly frustrated, nearly degenerate regimes present a greater challenge for finite‐depth QAOA. Hardware‐oriented tests are also performed in the non‐frustrated regime, where ideal QAOA simulations reproduce the exact result. On a noisy simulator and IonQ quantum hardware, however, the probability decreases as the number of layers increases, showing that hardware noise still limits the achievable performance of deeper QAOA circuits.