Reliable Graph Neural Network Explanations Through Adversarial Training.
Donald LovelandShusen LiuBhavya KailkhuraAnna M. HiszpanskiYong HanPublished in: CoRR (2021)
Keyphrases
- neural network
- training algorithm
- training process
- feed forward neural networks
- multi layer perceptron
- feedforward neural networks
- artificial neural networks
- training patterns
- back propagation
- neural network training
- backpropagation algorithm
- neural network structure
- recurrent networks
- training set
- graph structure
- train a neural network
- training phase
- graph model
- graph theoretic
- pattern recognition
- neural network model
- connected components
- structured data
- random walk
- bipartite graph
- graph theory
- fuzzy logic
- multi layer
- learning rate
- directed acyclic graph
- test set
- weighted graph
- graph representation
- activation function
- graph based algorithm
- graph databases
- fuzzy neural network
- hidden layer
- recurrent neural networks
- multilayer perceptron
- feed forward
- radial basis function
- supervised learning
- semi supervised
- multi agent
- clustering algorithm
- learning algorithm