… —Federated Learning (FL) is a promising machine learning … aggregation and model
synchronization among large-scale … a novel synchronization scheme named Partial …

… Federated Learning (FL) is a distributed learning paradigm with several challenges that
differentiate it from traditional distributed learning, … – they perform a synchronized aggregation of …

… exclude them from synchronization (but still update them locally), and only synchronize the
rest of the model to the central server. We find that such a partial synchronization method may …

… other schemes, which is caused by our partial synchronization and local update strategy. In
[4, … By allowing more GPUs to participate in partial synchronization, the training effect can be …

… goal in federated learning and has attracted attention from many researchers. Parameter
synchronization (… synchronization, ie, communication efficiency problem and straggler problem. …

… Abstract—Federated learning allows edge devices to collaboratively train a global …
synchronize the rest of the model to the central server. We find that such a partial synchronization …

… federated learning, which allows nodes to only synchronize the locally-trained models instead
their own original data. Conventional federated learning … , in real-world federated learning …

… We can easily find that, synchronized federated learning will achieve the best
convergence in terms of accuracy and speed. While, when the asynchronized straggler parameter …

… system based on permissioned blockchains that can effectively alleviate the overhead of
the synchronous federated learning algorithm on the synchronization problem, while the …

… In this work, we introduce a novel energy-efficient Semi-Synchronous Federated Learning
… We define the synchronization time period based on the data amounts and computational …