Quantum Neural Network for Quantum Neural Computing

By Min-Gang Zhou et al

Published on May 16, 2023

Read the original document by opening this link in a new tab.

Table of Contents

Introduction
Soft Quantum Neurons
Quantumness of Quantum Neurons
Soft Quantum Neural Network
Results

Summary

Neural networks have achieved impressive breakthroughs in both industry and academia. This paper introduces a new quantum neural network model for quantum neural computing, addressing the challenge of developing neural networks on quantum computing devices. The model utilizes single-qubit operations and measurements on real-world quantum systems, reducing implementation difficulties. It demonstrates exceptional nonlinear classification ability and robustness to noise, with applications in handwritten digit recognition. The proposed soft quantum neurons exhibit quantum correlations and entanglement, distinguishing them from classical neural models. The paper presents a fully-connected soft quantum feedforward neural network for supervised learning. The network operates probabilistically due to the measurement randomness, with stability achieved through multiple runs. The loss function, based on mean squared error, guides parameter updates for network optimization.