ENADPool: The Edge-Node Attention-based Differentiable Pooling for Graph Neural Networks.
Zhehan ZhaoLu BaiLixin CuiMing LiYue WangLixiang XuEdwin R. HancockPublished in: CoRR (2024)
Keyphrases
- undirected graph
- neural network
- edge weights
- weighted graph
- graph structure
- directed graph
- disjoint paths
- strongly connected
- densely connected
- spanning tree
- directed acyclic graph
- genetic algorithm
- pattern recognition
- approximation algorithms
- graph model
- edge information
- vertex set
- bipartite graph
- random walk
- nodes of a graph
- finding the shortest path
- artificial neural networks
- shortest path
- graph representation
- edge detection
- fuzzy logic
- root node
- path length
- self organizing maps
- tree structure
- complex networks
- connected components
- feed forward
- overlapping communities
- activation function
- recurrent neural networks
- edge map
- fault diagnosis
- neural network model
- multilayer perceptron
- neural nets