RSC: Accelerating Graph Neural Networks Training via Randomized Sparse Computations.
Zirui LiuShengyuan ChenKaixiong ZhouDaochen ZhaXiao HuangXia HuPublished in: CoRR (2022)
Keyphrases
- neural network
- training process
- training algorithm
- feedforward neural networks
- backpropagation algorithm
- feed forward neural networks
- back propagation
- multi layer perceptron
- weighted graph
- artificial neural networks
- structured data
- fuzzy logic
- multilayer perceptron
- graph theory
- graph representation
- random walk
- neural network training
- neural nets
- graph structure
- avoid overfitting
- training set
- adjacency matrix
- graph theoretic
- error back propagation
- genetic algorithm
- high dimensional
- pattern recognition
- graph mining
- multi layer
- sparse data
- recurrent neural networks
- self organizing maps
- fault diagnosis
- recurrent networks
- training samples
- elastic net
- supervised learning
- gaussian graphical models