Pipelined Backpropagation at Scale: Training Large Models without Batches.
Atli KossonVitaliy ChileyAbhinav VenigallaJoel HestnessUrs KösterPublished in: MLSys (2021)
Keyphrases
- back propagation
- training algorithm
- neural network
- feedforward neural networks
- artificial neural networks
- feed forward neural networks
- multilayer perceptron neural network
- neural nets
- backpropagation algorithm
- feed forward
- bp algorithm
- training data
- learning algorithm
- multilayer perceptron
- probabilistic model
- neural network ensemble
- training set
- hidden layer
- connection weights
- mlp neural networks