BNS-GCN: Efficient Full-Graph Training of Graph Convolutional Networks with Partition-Parallelism and Random Boundary Node Sampling.
Cheng WanYoujie LiAng LiNam Sung KimYingyan LinPublished in: MLSys (2022)
Keyphrases
- graph structure
- directed graph
- random walk
- overlapping communities
- edge weights
- graph structures
- vertex set
- bipartite graph
- undirected graph
- graph partitioning
- training set
- graph theory
- small world
- partitioning algorithm
- densely connected
- swendsen wang
- social networks
- small world networks
- random sampling
- bayesian networks