FedFa: A Fully Asynchronous Training Paradigm for Federated Learning

arXiv:2404.11015v2 Announce Type: replace
Abstract: Federated learning has been identified as an efficient decentralized training paradigm for scaling the machine learning model training on a large number of devices while guaranteeing the data privacy of the trainers. FedAvg has become a foundational parameter update strategy for federated learning, which has been promising to eliminate the effect of the heterogeneous data across clients and guarantee convergence. However, the synchronization parameter update barriers for each communication round during the training significant time …

abstract arxiv asynchronous become cs.ai cs.dc cs.lg data data privacy decentralized devices federated learning foundational machine machine learning machine learning model paradigm privacy scaling strategy training type update