PipeGCN: Efficient Full-Graph Training of Graph Convolutional Networks with Pipelined Feature Communication.
Cheng WanYoujie LiCameron R. WolfeAnastasios KyrillidisNam Sung KimYingyan LinPublished in: ICLR (2022)
Keyphrases
- graph structure
- graph theory
- random walk
- network structure
- directed graph
- graph theoretic
- graph representation
- vector representation
- pairwise
- graph model
- network analysis
- average degree
- social graphs
- fully connected
- graph partitioning
- undirected graph
- graph databases
- weighted graph
- computer networks
- structured data
- feature vectors