CATGNN: Cost-Efficient and Scalable Distributed Training for Graph Neural Networks.
Xin HuangWeipeng ZhuoMinh Phu VuongShiju LiJongryool KimBradley ReesChul-Ho LeePublished in: CoRR (2024)
Keyphrases
- cost efficient
- scalable distributed
- neural network
- training process
- training algorithm
- multi layer perceptron
- feedforward neural networks
- neural network training
- backpropagation algorithm
- training set
- supervised learning
- feed forward neural networks
- graph structure
- back propagation
- directed acyclic graph
- genetic algorithm
- training phase
- neural network structure
- weighted graph
- graph theory
- pattern recognition
- directed graph
- structured data
- recurrent neural networks
- artificial neural networks
- self organizing maps
- graph representation
- training examples
- error back propagation
- support vector
- pairwise
- recurrent networks
- online learning
- load balancing
- graph model
- feed forward
- file system
- bipartite graph