Subgraph Federated Learning for Local Generalization
This addresses the challenge of generalization in federated graph learning for privacy-preserving collaborative training, but appears incremental as it builds on existing methods to mitigate a specific bottleneck.
The paper tackles the problem of local overfitting in federated learning on graphs due to mutable data and label distribution shifts, and proposes FedLoG which uses global synthetic data to enhance local model generalization, outperforming baselines in experimental settings.
Federated Learning (FL) on graphs enables collaborative model training to enhance performance without compromising the privacy of each client. However, existing methods often overlook the mutable nature of graph data, which frequently introduces new nodes and leads to shifts in label distribution. Since they focus solely on performing well on each client's local data, they are prone to overfitting to their local distributions (i.e., local overfitting), which hinders their ability to generalize to unseen data with diverse label distributions. In contrast, our proposed method, FedLoG, effectively tackles this issue by mitigating local overfitting. Our model generates global synthetic data by condensing the reliable information from each class representation and its structural information across clients. Using these synthetic data as a training set, we alleviate the local overfitting problem by adaptively generalizing the absent knowledge within each local dataset. This enhances the generalization capabilities of local models, enabling them to handle unseen data effectively. Our model outperforms baselines in our proposed experimental settings, which are designed to measure generalization power to unseen data in practical scenarios. Our code is available at https://github.com/sung-won-kim/FedLoG