Training Fair Models in Federated Learning without Data Privacy Infringement | IEEE Conference Publication | IEEE Xplore