DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks.
Pál András PappKarolis MartinkusLukas FaberRoger WattenhoferPublished in: NeurIPS (2021)
Keyphrases
- neural network
- pattern recognition
- directed graph
- artificial neural networks
- structured data
- graph representation
- graph based algorithm
- fuzzy logic
- graph structure
- connected components
- graph theoretic
- graph theory
- dependency graph
- self organizing maps
- expressive power
- neural network model
- graph matching
- multi layer
- graph clustering
- random walk
- data sets
- undirected graph
- stable set
- graph structures
- random graphs
- spanning tree
- graph mining
- training process
- fuzzy systems
- bipartite graph
- social network analysis
- search algorithm
- clustering algorithm