Adversarial Weight Perturbation Improves Generalization in Graph Neural Networks.
Yihan WuAleksandar BojchevskiHeng HuangPublished in: AAAI (2023)
Keyphrases
- neural network
- pattern recognition
- graph theory
- graph structure
- genetic algorithm
- random walk
- weighted graph
- graph model
- edge weights
- directed graph
- bipartite graph
- self organizing maps
- fuzzy logic
- artificial neural networks
- multi agent
- spanning tree
- graph partitioning
- fuzzy systems
- weight update
- multilayer perceptron
- graph clustering
- connected components
- undirected graph
- back propagation
- graph representation
- data mining
- neural nets
- dependency graph
- rule extraction
- decision trees
- graph theoretic
- weighting scheme
- neural network model
- graph mining
- link analysis
- feed forward
- multi layer
- graph matching
- recurrent neural networks