GNN-VPA: A Variance-Preserving Aggregation Strategy for Graph Neural Networks.
Lisa SchneckenreiterRichard FreinschlagFlorian SestakJohannes BrandstetterGünter KlambauerAndreas MayrPublished in: Tiny Papers @ ICLR (2024)
Keyphrases
- neural network
- graph structure
- graph representation
- neural nets
- graph theoretic
- fuzzy systems
- graph theory
- bipartite graph
- data sets
- recurrent neural networks
- pattern recognition
- artificial neural networks
- directed graph
- fuzzy logic
- graph matching
- graph data
- graph model
- weighted graph
- random walk
- self organizing maps
- neural network model
- search strategy
- correlation coefficient
- data aggregation
- variance reduction
- dependency graph
- graph based algorithm