SmartSAGE: Training Large-scale Graph Neural Networks using In-Storage Processing Architectures.
Yunjae LeeJinha ChungMinsoo RhuPublished in: CoRR (2022)
Keyphrases
- neural network
- training process
- training algorithm
- feedforward neural networks
- processing capabilities
- pattern recognition
- multi layer perceptron
- storage and retrieval
- feed forward neural networks
- back propagation
- random access
- memory management
- random walk
- data processing
- neural network training
- real world
- information processing
- graph theory
- real time
- graph structure
- graph representation
- backpropagation algorithm
- graph matching
- directed graph
- fuzzy logic
- structured data
- error back propagation
- neural architectures
- file system
- graph model
- artificial neural networks
- neural network model
- graph theoretic
- genetic algorithm
- training examples
- self organizing maps
- connected components
- data storage
- graph partitioning
- storage requirements
- directed acyclic graph
- bipartite graph