<ul><li>Mobile and Web-of-Things (WoT) devices generate vast amounts of data for machine learning applications.</li><li>Federated Learning (FL) allows clients to collaboratively train a shared model without transferring private data.</li><li>Existing FL methods prioritize either global generalization or local personalization, limiting the potential of diverse client data.</li><li>The proposed FedRIR framework enhances global generalization and local personalization by rethinking information representation.</li></ul>

FedRIR: Rethinking Information Representation in Federated Learning

Discover more