Learning from straggler clients in federated learning
This addresses the challenge of handling delayed updates in federated learning, which is crucial for real-world applications with heterogeneous client devices, though it appears incremental as it builds on existing methods.
The paper tackled the problem of learning from severely delayed clients in federated learning, finding that existing algorithms struggle with this issue. It introduced two new algorithms, FARe-DUST and FeAST-on-MSG, which outperformed existing ones in accuracy for straggler clients and provided better trade-offs between training time and total accuracy on benchmarks like EMNIST, CIFAR-100, and StackOverflow.
How well do existing federated learning algorithms learn from client devices that return model updates with a significant time delay? Is it even possible to learn effectively from clients that report back minutes, hours, or days after being scheduled? We answer these questions by developing Monte Carlo simulations of client latency that are guided by real-world applications. We study synchronous optimization algorithms like FedAvg and FedAdam as well as the asynchronous FedBuff algorithm, and observe that all these existing approaches struggle to learn from severely delayed clients. To improve upon this situation, we experiment with modifications, including distillation regularization and exponential moving averages of model weights. Finally, we introduce two new algorithms, FARe-DUST and FeAST-on-MSG, based on distillation and averaging, respectively. Experiments with the EMNIST, CIFAR-100, and StackOverflow benchmark federated learning tasks demonstrate that our new algorithms outperform existing ones in terms of accuracy for straggler clients, while also providing better trade-offs between training time and total accuracy.