The Surprising Power of Graph Neural Networks with Random Node Initialization.
Ralph AbboudIsmail Ilkan CeylanMartin GroheThomas LukasiewiczPublished in: IJCAI (2021)
Keyphrases
- neural network
- graph structure
- directed graph
- undirected graph
- graph model
- random walk
- structured data
- genetic algorithm
- root node
- pattern recognition
- overlapping communities
- edge weights
- graph theory
- fuzzy logic
- graph theoretic
- graph representation
- path length
- graph partitioning
- finding the shortest path
- nodes of a graph
- bipartite graph
- neural nets
- power consumption
- back propagation
- training process
- recurrent neural networks
- feed forward
- neural network model
- connected components
- tree structure
- evolutionary algorithm
- artificial neural networks