Co-authored by Xinchi Huang, Taichi Kosugi, Hirofumi Nishi, and Yu-ichiro Matsushita of Quemix, the paper has addressed the crucial task of optimizing circuit depth and gate count in the NISQ era, this study focuses on the implementation of specific diagonal unitary matrices with reflection symmetry. It demonstrates the further simplification of existing quantum circuits, and proposes an algorithm to optimize the number of entanglement gates. As a result, it is shown that compared to previous methods for synthesizing diagonal unitary matrices, both gate count and circuit depth can be reduced by nearly half.
Summary:
In the context of the NISQ era, optimizing quantum circuits, especially those with entanglement gates like the CNOT gate, is critical. Diagonal unitary matrices are a special class among all unitary operators, crucial in many quantum algorithms. Previous research focused on quantum circuits for general diagonal unitary matrices using the {CNOT, Rz} gate set and proposed optimal synthesis methods in terms of circuit depth.
This paper emphasizes the implementation of a specific category of diagonal unitary matrices with reflection symmetry. These matrices hold promise for applications such as real-time evolution of first quantized Hamiltonians for interacting particles using quantum circuits. Leveraging their symmetric properties, the study demonstrates further simplification of existing quantum circuits and introduces a constructive algorithm for optimizing (entanglement) gate count. Compared to prior methods for general diagonal unitary matrices, our proposed algorithm achieves nearly a 50% reduction in both gate count and circuit depth.
This research demonstrates its significant contribution to the improvement in NISQ device efficiency through the optimization of entanglement gates and simplification of quantum circuits. It also showcases the efficient design of quantum circuits for enabling real-time evolution of one-dimensional Hamiltonians for interacting particles, which is a substantial step toward practical applications in quantum computing.
This technology accelerates the simulation of real-time dynamics on quantum computers.
