Efficient Random Quantum Circuit Generator: A Benchmarking Approach for Quantum Simulators

Recently, many proposed simulators have been proposed. However, they lack a standard evaluation, including a benchmarking dataset and metrics set, leading to low confidence. To solve this problem, in this research, we propose a general random quantum circuit (RQC) generator. The proposed generator consumes fewer resources than the Haar random generator. When increasing the number of qubits, the resources used by the RQC generator remain constant, while the resources for Haar increase exponentially. To demonstrate the applicability of RQC, we evaluated on 14GB dataset of 15,000 generated quantum circuits through four simulators and three hardware-based simulators. Moreover, the evaluation metrics have been clarified for the benchmarking process of quantum simulators. The experiment results based on these metrics demonstrate that the RQC generator is a practical component for random quantum circuit benchmarking.