POWER is a federated continual graph learning framework that combines replay of carefully selected old nodes with server-side prototype reconstruction and trajectory-weighted knowledge transfer, outperforming adapted baselines on eight graph datasets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Federated Continual Graph Learning
POWER is a federated continual graph learning framework that combines replay of carefully selected old nodes with server-side prototype reconstruction and trajectory-weighted knowledge transfer, outperforming adapted baselines on eight graph datasets.