GNN-FiLM: Graph Neural Networks with Feature-wise Linear Modulation.
Marc BrockschmidtPublished in: ICML (2020)
Keyphrases
- neural network
- graph theory
- graph representation
- connected components
- genetic algorithm
- pattern recognition
- artificial neural networks
- random walk
- linear features
- preprocessing stage
- pairwise
- directed graph
- image features
- vector representation
- feature vectors
- graph structure
- amplitude modulation
- highly non linear
- series parallel
- graph partitioning
- spanning tree
- graph databases
- directed acyclic graph
- multi layer
- bipartite graph
- neural nets
- graph matching
- feed forward
- structured data
- feature set
- fuzzy logic