Privacy-Preserving Federated Graph Neural Network Learning on Non-IID Graph Data

Algorithm 1

The federated DeepWalk framework.

Input:: the set of clients

  G_i: the local subgraph hold by c_i

  U_k: the public nodes shared among C

Output: the matrix of node representation

  of G_i

 1: LOCAL CLIENTS:

 2: for each client c_i ∈ C do

 3: Compute the DeepWalk model weights Φ_i

 4: Generate the public nodes’ embeddings X_i of U_k

   from Φ_i:

 5:  

 6: Upload X_i to the server

 7: end for

 8:

 9: while not converge do

 10: SERVER:

 11: for each i ∈ kdo

 12:  for each do

 13:   Align X_j into c_i’s space:

 14:  end for

 15:  Aggregate all the aligned embeddings with

 16:   

 17:  distribute to client c_i for local update

 18: end for

 19:

 20: LOCAL CLIENTS:

 21: for each client c ∈ C do

 22:  Substitute the public nodes’ embeddings in Φ_i

    by

 23:   

 24:  Initial the DeepWalk model with

 25:  Compute the model weights Φ_i

 26: end for

 27: end while

 28: return the matrix of node representation

  of Gi