Attention-Augmented Parametric Kernel Graph Neural Network (APKGNN) for Node Classification.
Avishek BoseWilliam H. HsuPublished in: ICMLA (2023)
Keyphrases
- neural network
- learning vector quantization
- graph structure
- pattern recognition
- support vector
- feature space
- classification accuracy
- decision trees
- pattern classification
- machine learning
- multi layer perceptron
- kernel machines
- binary decision tree
- kernel density estimators
- graph theory
- directed graph
- text classification
- feature selection
- feature extraction
- kernel function
- random walk
- density estimation
- back propagation
- multiple kernel learning
- support vector machine
- kernel fisher discriminant
- training data
- artificial neural networks
- path length
- feature vectors
- undirected graph
- training process
- kernel methods
- classification method
- tree structure
- svm classifier
- structured data