WHERE to Generate Matters: Budget-Aware Synthetic Augmentation for Label Skewed Federated Learning
For federated learning practitioners, FedEAS provides a budget-aware augmentation policy that drastically reduces computation cost while maintaining accuracy.
FedEAS reduces the synthetic data generation budget by 94.1% while recovering most accuracy gains of full class balancing in label-skewed federated learning, outperforming uniform allocation by up to 18.82% on CIFAR-10/100.
Label skew in federated learning (FL) causes client drift and degrades global accuracy. Synthetic data augmentation can reduce this imbalance; however, full class balancing requires substantial computation cost. We propose FedEAS, a policy that assigns each client an entropy-adaptive per-class generation budget computed from its local label distribution. The budget jointly decides \emph{how much} each client generates and \emph{WHERE} the samples go. Accordingly, the total generation budget follows from the per-client budgets rather than being fixed in advance. FedEAS recovers most of the accuracy gain of full class balancing while reducing the generation budget by 94.1\%. At the same total generation budget, it outperforms Uniform allocation by up to 18.82\% across CIFAR-10 and CIFAR-100.