Comparing concepts of quantum and classical neural network models for image classification task

TL;DR

This paper compares classical and quantum neural networks for image classification, demonstrating that quantum models can outperform classical ones in accuracy and convergence despite simulation challenges.

Contribution

It introduces a hybrid quantum-classical neural network for image classification and provides a comparative analysis showing quantum advantages over classical models.

Findings

Quantum neural networks achieve higher accuracy than classical models.

Quantum models show better convergence during training.

Simulation of quantum networks is time-consuming but feasible.

Abstract

While quantum architectures are still under development, when available, they will only be able to process quantum data when machine learning algorithms can only process numerical data. Therefore, in the issues of classification or regression, it is necessary to simulate and study quantum systems that will transfer the numerical input data to a quantum form and enable quantum computers to use the available methods of machine learning. This material includes the results of experiments on training and performance of a hybrid quantum-classical neural network developed for the problem of classification of handwritten digits from the MNIST data set. The comparative results of two models: classical and quantum neural networks of a similar number of training parameters, indicate that the quantum network, although its simulation is time-consuming, overcomes the classical network (it has better convergence and achieves higher training and testing accuracy).