Adversarial Training on Weights for Graph Neural Networks.
Haotian XueXin WangYing WangPublished in: ACAI (2022)
Keyphrases
- neural network
- training process
- feedforward neural networks
- training algorithm
- multi layer perceptron
- hidden neurons
- feed forward neural networks
- training speed
- backpropagation algorithm
- edge weights
- back propagation
- linear combination
- graph theory
- random walk
- graph theoretic
- graph partitioning
- error back propagation
- graph representation
- training phase
- number of hidden units
- weighted graph
- multi agent
- pattern recognition
- neural network model
- genetic algorithm
- linearly combined
- connection weights
- fuzzy logic
- bipartite graph
- graph structure
- weighted sum
- connected components
- fault diagnosis
- test set
- training examples
- spanning tree
- artificial neural networks
- network architecture
- graph model
- training data
- graph mining
- directed acyclic graph