Federated Learning with Hypergradient-based Online Update of Aggregation Weights
For federated learning practitioners, it addresses the need for adaptability to heterogeneous data and varying communication conditions with low computational overhead.
FedHAW proposes online updates of aggregation weights in federated learning using hypergradients, achieving high generalization performance in heterogeneous environments and robustness to communication errors.
Federated learning using mobile and Internet of Things devices requires not only the ability to handle heterogeneity of clients' data distributions but also high adaptability to varying communication environments. We propose FedHAW (Federated Learning with Hypergradient-based update of Aggregation Weights) that implements online updates of aggregation weights. FedHAW updates the aggregation weights by using hypergradient, the gradient of the objective function with respect to the weights, which can be calculated with low computational overhead. Simulation results show that the proposed method possesses high generalization performance in heterogeneous environments and high robustness to communication errors.