PP-GNN: Pretraining Position-aware Graph Neural Networks with the NP-hard metric dimension problem.
Michael SunPublished in: Neurocomputing (2023)
Keyphrases
- neural network
- np hard
- artificial neural networks
- maximum weight
- pattern recognition
- graph theory
- special case
- genetic algorithm
- graph theoretic
- graph structure
- bipartite graph
- scheduling problem
- fuzzy logic
- back propagation
- directed graph
- lower bound
- graph representation
- graph model
- directed acyclic graph
- multi layer
- neural network model
- metric space
- weighted graph
- optimal solution
- computational complexity
- branch and bound algorithm
- neighborhood structure
- evaluation metrics
- recurrent neural networks
- worst case
- connected components
- self organizing maps
- random walk
- np complete
- linear programming
- maximum common subgraph
- np hardness
- computationally hard
- spanning tree
- graph databases
- multilayer perceptron
- feed forward
- constraint satisfaction problems