Encoding a quantum algorithm into a quantum circuit, under the constraints of qubit counts, connectivities and coherence times, is an essential step for making the best use of near-term quantum devices. We approach this “circuit optimization” task with a multi-tiered quantum circuit optimization protocol called AQCEL. Two separate ideas constitute the implementation of AQCEL protocol: recognition of repeated patterns of quantum gates, and reduction of quantum gates with the identification of computational basis states. A novelty resides in the AQCEL functionality of performing optimization on a quantum device using polynomial resources.
The AQCEL optimization is deployed on a quantum algorithm in high-energy physics simulation, demonstrating a drastic reduction of the circuit length while maintaining the computational accuracy. We plan to extend the AQCEL protocol with optimized control/synthesis of all microwave-based quantum gates for near-term devices, including the possibility of using higher-energy states for information storage.