Distributed Quantum Machine Learning: Federated and Model-Parallel Approaches

In this article, we explore two types of distributed quantum machine learning (DQML) methodologies: quantum federated learning and quantum model-parallel learning. We discuss the challenges encountered in DQML, propose potential solutions, and highlight future research directions in this rapidly evolving field. Additionally, we implement two solutions tailored to the two types of DQML, aiming to enhance the reliability of the computing process. Our results show the potential of DQML in the current Noisy Intermediate-Scale Quantum era.

Distributed quantum machine learning (DQML) is an emerging field that combines quantum machine learning (QML) with distributed computing. QML utilizes distinctive quantum mechanics properties, like superposition and entanglement, to potentially enhance traditional machine learning algorithms. However, the current stage of quantum computing technology, often referred to as the Noisy Intermediate-Scale Quantum (NISQ) era, imposes limitations on the size and complexity of QML models implemented with variational quantum circuits (VQCs). These constraints can restrict the performance and applicability of QML methods. To address these challenges, integrating QML with distributed computing has emerged as a strategic and forward-looking approach. This hybrid approach aims to overcome the individual limitations of quantum devices by harnessing distributed quantum computing’s power to manage and process complex tasks across multiple quantum computing nodes, thus amplifying the capabilities of QML models.