Data classification and forecasting with Quantum Machine Learning

Speaker: Luca Asproni (Data Reply)

Abstract

Machine learning is an application of Artificial Intelligence, based on computer algorithms able to learn from data without being explicitly programmed.

These algorithms have proven their strength in the last years, with many applications in different contexts.

Yet, in some cases they could suffer of accuracy or performances issues. In order to overcome this potential limitations, a new approach arises in the modern technological landscape: the Quantum Machine Learning (QML), which is a hybrid approach mixing Quantum computing with classical algorithms.

In this work we report our first findings in three different classes of QML algorithms: a Quantum SVM based on both a variational and kernel approach and a specific Quantum Neural Network (QNN) implementation.

In this work we show how this kind of algorithms can be applied in different scenarios and what are the current limitations and future perspectives.