Referring to Section 4.3, FedSage+ includes two phases. Firstly, all data owners in the distributed subgraph system jointly train NeighGen models through sharing gradients. Next, after every local graph mended with synthetic neighbors generated by the respective NeighGen model, the system executes FedSage to obtain the generalized node classification model. Algorithm 1 shows the pseudo code for FedSage+. To perform node classification on G, we consider a GNNF with K aggregation operations1 and each aggregation operation contains Rfully-connected layers.

For notation simplicity, GNNF here is considered in GNTK format. The weights ofF, φ is i.i.d. Let ha,bi denote inner-product of vectoraandb. Next,weexperimentally verify the necessity of training locally specialized NeighGen. Specifically, we present theL2 distance between the averaged feature distributions of neighborhoods from these three types of graphs to show how the NeighGen generated missing neighborsnarrowthegap.

Graphs have been widely used in data mining and machine learning due to their unique representation of real-world objects and their interactions.