Calibrate and Debias Layer-wise Sampling for Graph Convolutional Networks.
Yifan ChenTianning XuDilek Hakkani-TurDi JinYun YangRuoqing ZhuPublished in: Trans. Mach. Learn. Res. (2023)
Keyphrases
- graph representation
- dynamic networks
- random walk
- average degree
- social networks
- highly connected
- small world
- pairwise
- heterogeneous networks
- graph structures
- graph structure
- community discovery
- monte carlo
- structured data
- graph theoretic
- fully connected
- restricted boltzmann machine
- social graphs
- edge weights
- directed edges
- graph partitioning
- graph model
- complex networks
- weighted graph
- graph matching
- unsupervised learning
- wireless networks
- sample size
- overlapping communities
- citation networks
- connected components
- directed graph
- graph theory
- application layer
- network analysis
- sampling algorithm
- network architecture