Federated Learning of Large Models at the Edge via Principal Sub-Model Training

Open in new window