FedFed: Feature Distillation against Data Heterogeneity in Federated Learning

Federated learning (FL) typically faces data heterogeneity, i.e., distribution shifting among clients. Sharing clients’ information has shown great potentiality in mitigating data heterogeneity, yet incurs a dilemma in preserving privacy and promoting model performance. To alleviate the dilemma, we raise a fundamental question: $\mathit{\text{Is}}\mathit{\text{it}}\mathit{\text{possible}}\mathit{\text{to}}\mathit{\text{share}}\mathit{\text{partial}}\mathit{\text{features}}\mathit{\text{in}}\mathit{\text{the}}\mathit{\text{data}}\mathit{\text{to}}\mathit{\text{tackle}}\mathit{\text{data}}\mathit{\text{heterogeneity?}}$ In this work, we give an affirmative answer to this question by proposing a novel approach called {$\mathbf{\text{Fed}}$erated $\mathbf{\text{Fe}}$ature $\mathbf{\text{d}}$istillation} (FedFed). Specifically, FedFed partitions data into performance-sensitive features (i.e., greatly contributing to model performance) and performance-robust features (i.e., limitedly contributing to model performance). The performance-sensitive features are globally shared to mitigate data heterogeneity, while the performance-robust features are kept locally. FedFed enables clients to train models over local and shared data. Comprehensive experiments demonstrate the efficacy of FedFed in promoting model performance.