Interpreting Unfairness in Graph Neural Networks via Training Node Attribution.
Yushun DongSong WangJing MaNinghao LiuJundong LiPublished in: AAAI (2023)
Keyphrases
- neural network
- graph structure
- directed graph
- training process
- training algorithm
- pattern recognition
- graph representation
- multi layer perceptron
- feedforward neural networks
- feed forward neural networks
- neural network training
- random walk
- backpropagation algorithm
- overlapping communities
- back propagation
- neural network structure
- undirected graph
- artificial neural networks
- finding the shortest path
- root node
- path length
- graph model
- directed acyclic graph
- graph theory
- fuzzy logic
- strongly connected
- recurrent neural networks
- training examples
- error back propagation
- training phase
- training set
- edge weights
- graphical models
- multi layer
- connected components
- structured data