FedSSP: Federated Graph Learning with Spectral Knowledge and Personalized Preference
This work addresses privacy-preserving and personalized graph learning for non-IID participants in federated settings, representing an incremental improvement over prior methods.
The paper tackles structural heterogeneity in personalized federated graph learning by sharing generic spectral knowledge and using personalized preference modules, achieving state-of-the-art performance in cross-dataset and cross-domain experiments.
Personalized Federated Graph Learning (pFGL) facilitates the decentralized training of Graph Neural Networks (GNNs) without compromising privacy while accommodating personalized requirements for non-IID participants. In cross-domain scenarios, structural heterogeneity poses significant challenges for pFGL. Nevertheless, previous pFGL methods incorrectly share non-generic knowledge globally and fail to tailor personalized solutions locally under domain structural shift. We innovatively reveal that the spectral nature of graphs can well reflect inherent domain structural shifts. Correspondingly, our method overcomes it by sharing generic spectral knowledge. Moreover, we indicate the biased message-passing schemes for graph structures and propose the personalized preference module. Combining both strategies, we propose our pFGL framework FedSSP which Shares generic Spectral knowledge while satisfying graph Preferences. Furthermore, We perform extensive experiments on cross-dataset and cross-domain settings to demonstrate the superiority of our framework. The code is available at https://github.com/OakleyTan/FedSSP.