Quantum optical model of an artificial neuron

TL;DR

This paper introduces two quantum circuit synthesis algorithms for implementing a quantum neuron, including a quantum optical variant that reduces resource requirements, supported by simulations demonstrating their effectiveness.

Contribution

It presents novel quantum circuit synthesis algorithms for quantum neurons and proposes a quantum optical model that lowers quantum resource usage.

Findings

Quantum circuit synthesis algorithms successfully implement quantum neurons.

Quantum optical variant reduces quantum resource requirements.

Simulations validate the effectiveness of the proposed models.

Abstract

Magnini \emph{et al.} [\emph{Mach. Learn.: Sci. Technol. 1 (2020) 045008}] recently introduced a qubit-based model of an artificial neuron, along with its applications. The design of its quantum circuit is pivotal for effective implementation. In this context, we present two quantum circuit synthesis algorithms tailored for the realisation of the quantum neuron. Comprehensive circuit simulations are conducted, and the resulting performance is assessed using the circuit cost metric. Additionally, we propose a quantum optical variant of the qubit-based quantum neuron, which offers a reduction in quantum resource requirements. To substantiate this, we introduce a quantum optical circuit synthesis algorithm and validate its efficacy through numerical simulations of prototype models.