Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks.
Yujun YanMilad HashemiKevin SwerskyYaoqing YangDanai KoutraPublished in: ICDM (2022)
Keyphrases
- convolutional neural networks
- random walk
- graph theory
- graph representation
- structured data
- graph based algorithm
- graph model
- convolutional network
- graph theoretic
- connected components
- bipartite graph
- directed graph
- graph clustering
- graph search
- planar graphs
- multiscale
- graph mining
- graph matching
- minimum spanning tree
- graph construction
- adjacency matrix
- data mining