A native measurement-based QAOA algorithm, applied to the MAX K-CUT problem (2304.03576v1)

Abstract: Photonic quantum computers, programmed within the framework of the measurement-based quantum computing (MBQC), currently concur with gate-based platforms in the race towards useful quantum advantage, and some algorithms emerged as main candidates to reach this goal in the near term. Yet, the majority of these algorithms are only expressed in the gate-based model of computation, which is incompatible with photonic platforms. Methods to translate gate-based algorithms into the MBQC framework exist, but they are not always optimal in terms of resource cost. In our work, we propose an MBQC algorithm to run the Quantum Approximate Optimization Algorithm (QAOA). Furthermore, we apply the MBQC-QAOA algorithm to the MAX $K$-CUT problem, working for all values of $K$, expressing the cost Hamiltonian and its constraints in a form easily implementable in the MBQC model. We conclude analyzing the resource-cost of our algorithm, compared to the case of translating a gate-based QAOA algorithm into MBQC rules showing up to a 30-fold improvement. With our work, we contribute to close the gap between gate-based and MBQC near-term algorithms, a gap not reflecting the current status of the hardware development.