Asyn2F: An Asynchronous Federated Learning Framework with Bidirectional Model Aggregation

arXiv:2403.01417v1 Announce Type: new
Abstract: In federated learning, the models can be trained synchronously or asynchronously. Many research works have focused on developing an aggregation method for the server to aggregate multiple local models into the global model with improved performance. They ignore the heterogeneity of the training workers, which causes the delay in the training of the local models, leading to the obsolete information issue. In this paper, we design and develop Asyn2F, an Asynchronous Federated learning Framework with …

abstract aggregation arxiv asynchronous cs.dc cs.lg federated learning framework global multiple performance research server training type workers