FedFST: Mitigating Spectral Catastrophic Forgetting in Federated Graph Continual Learning
FedFST is a spectral-aware framework that mitigates dual spectral forgetting in Federated Graph Continual Learning, and comprises Historical High-Frequency Knowledge Restoration to reconstruct and preserve high-frequency inconsistency knowledge, and Historical Low-Frequency Semantic Transfer to stabilize low-frequency...