Bayesian Parameterized Quantum Circuit Optimization (BPQCO): A task and hardware-dependent approach

Abstract

Variational quantum algorithms (VQA) have emerged as a promising quantum alternative for solving optimization and machine learning problems using parameterized quantum circuits (PQCs). The design of these circuits influences the ability of the algorithm to efficiently explore the solution space and converge to more optimal solutions. Choosing an appropriate circuit topology, gate set, and parameterization scheme is determinant to achieve good performance. In addition, it is not only problem-dependent, but the quantum hardware used also has a significant impact on the results. Therefore, we present BPQCO, a Bayesian Optimization-based strategy to search for optimal PQCs adapted to the problem to be solved and to the characteristics and limitations of the chosen quantum hardware. To this end, we experimentally demonstrate the influence of the circuit design on the performance obtained for two classification problems (a synthetic dataset and the well-known Iris dataset), focusing on the design of the circuit ansatz. In addition, we study the degradation of the obtained circuits in the presence of noise when simulating real quantum computers. To mitigate the effect of noise, two alternative optimization strategies based on the characteristics of the quantum system are proposed. The results obtained confirm the relevance of the presented approach and allow its adoption in further work based on the use of PQCs.