Chimera: efficiently training large-scale neural networks with bidirectional pipelines.
Shigang LiTorsten HoeflerPublished in: SC (2021)
Keyphrases
- neural network
- training process
- training algorithm
- multi layer perceptron
- feed forward neural networks
- feedforward neural networks
- back propagation
- error back propagation
- small scale
- fuzzy logic
- training patterns
- artificial neural networks
- pattern recognition
- backpropagation algorithm
- neural network training
- real world
- feed forward
- data sets
- neural network structure
- recurrent neural networks
- neural network model
- real life
- genetic algorithm
- neural nets
- test set
- training samples
- training phase
- online learning
- data management
- supervised learning
- information systems
- learning algorithm
- database