<ul><li>A new model-driven quantum federated learning algorithm (mdQFL) has been introduced to address challenges in quantum federated learning (QFL).</li><li>The algorithm aims to tackle training bottlenecks, involvement of a large number of devices, and non-IID data distributions efficiently.</li><li>Through extensive experiments in the Qiskit environment, the mdQFL framework demonstrated a nearly 50% decrease in total communication costs while maintaining or exceeding model accuracy.</li><li>The experimental evaluation also includes a theoretical analysis of the proposed mdQFL algorithm and its complexities.</li></ul>

Communication Efficient Adaptive Model-Driven Quantum Federated Learning

Discover more