Phase polynomials synthesis algorithms for NISQ architectures and beyond

TL;DR

This paper introduces new algorithms for synthesizing phase polynomials tailored for NISQ quantum architectures, achieving lower CNOT counts and depths, with flexible trade-offs between circuit complexity and execution time.

Contribution

The paper presents novel synthesis algorithms for phase polynomials that outperform existing methods in efficiency and flexibility for NISQ devices.

Findings

Lower CNOT count and depth compared to state-of-the-art algorithms

Algorithms applicable to both full and partial connectivity architectures

Trade-off methods between circuit complexity and execution time

Abstract

We present a framework for the synthesis of phase polynomials that addresses both cases of full connectivity and partial connectivity for NISQ architectures. In most cases, our algorithms generate circuits with lower CNOT count and CNOT depth than the state of the art or have a significantly smaller running time for similar performances. We also provide methods that can be applied to our algorithms in order to trade an increase in the CNOT count for a decrease in execution time, thereby filling the gap between our algorithms and faster ones.