Quantum ensemble of trained classifiers

TL;DR

This paper explores how adding an optimization step to quantum ensembles of classifiers enhances their performance, leveraging superposition to create large, trainable classifier ensembles in quantum machine learning.

Contribution

It introduces an optimization-enhanced approach to quantum ensembles, demonstrating improved results over previous optimization-free methods.

Findings

Optimization improves classification accuracy.

Quantum ensembles can be effectively trained with added optimization.

Experimental results on benchmark datasets confirm performance gains.

Abstract

Through superposition, a quantum computer is capable of representing an exponentially large set of states, according to the number of qubits available. Quantum machine learning is a subfield of quantum computing that explores the potential of quantum computing to enhance machine learning algorithms. An approach of quantum machine learning named quantum ensembles of quantum classifiers consists of using superposition to build an exponentially large ensemble of classifiers to be trained with an optimization-free learning algorithm. In this work, we investigate how the quantum ensemble works with the addition of an optimization method. Experiments using benchmark datasets show the improvements obtained with the addition of the optimization step.