Noise-Adaptive Quantum Compilation Strategies Evaluated with Application-Motivated Benchmarks

TL;DR

This paper introduces a new noise-adaptive quantum compilation strategy optimized for heavy-hexagon lattice devices, demonstrating improved performance on deep and square circuits through application-motivated benchmarks.

Contribution

A novel, computationally efficient noise-adaptive compilation strategy tailored for heavy-hexagon lattice quantum devices, validated with application-motivated benchmarks.

Findings

Effective for deep circuits

Improves performance on square circuits

Outperforms some state-of-the-art approaches

Abstract

Quantum compilation is the problem of translating an input quantum circuit into the most efficient equivalent of itself, taking into account the characteristics of the device that will execute the computation. Compilation strategies are composed of sequential passes that perform placement, routing and optimization tasks. Noise-adaptive compilers do take the noise statistics of the device into account, for some or all passes. The noise statics can be obtained from calibration data, and updated after each device calibration. In this paper, we propose a novel noise-adaptive compilation strategy that is computationally efficient. The proposed strategy assumes that the quantum device coupling map uses a heavy-hexagon lattice. Moreover, we present the application-motivated benchmarking of the proposed noise-adaptive compilation strategy, compared with some of the most advanced state-of-art approaches. The presented results seem to indicate that our compilation strategy is particularly effective for deep circuits and for square circuits.