MaxK-GNN: Towards Theoretical Speed Limits for Accelerating Graph Neural Networks Training.
Hongwu PengXi XieKaustubh ShivdikarMd Amit HasanJiahui ZhaoShaoyi HuangOmer KhanDavid KaeliCaiwen DingPublished in: CoRR (2023)
Keyphrases
- neural network
- training process
- training algorithm
- multi layer perceptron
- feedforward neural networks
- feed forward neural networks
- high speed
- backpropagation algorithm
- neural network training
- training speed
- random walk
- graph representation
- directed graph
- training patterns
- back propagation
- error back propagation
- graph theory
- training set
- multi layer
- training data
- structured data
- artificial neural networks
- graph theoretic
- multilayer neural network
- neural nets
- graph matching
- genetic algorithm
- neural network model
- pattern recognition
- supervised learning
- connected components
- self organizing maps
- real time
- neural network structure
- online learning
- theoretical analysis
- training samples
- training examples
- test set
- bipartite graph
- graph partitioning
- training phase
- spanning tree
- directed acyclic graph
- weighted graph