FedGCN: Convergence and Communication Tradeoffs in Federated Training of Graph Convolutional Networks.
Yuhang YaoCarlee Joe-WongPublished in: CoRR (2022)
Keyphrases
- high bandwidth
- computer networks
- graph representation
- graph theory
- average degree
- random walk
- graph model
- graph structure
- social networks
- small world
- graph theoretic
- packet switched
- community discovery
- graph structures
- training set
- echo state networks
- network analysis
- graph layout
- convergence rate
- weighted graph
- communication networks
- interconnection networks
- supervised learning
- directed graph
- betweenness centrality
- network size
- overlapping communities
- recurrent networks
- neural network
- hearing impaired
- restricted boltzmann machine
- dynamic networks
- biological networks
- directed acyclic graph
- training process
- communication cost
- bipartite graph
- graph matching
- complex networks
- network structure
- distributed systems