Multiplexed gradient descent: Fast online training of modern datasets on hardware neural networks without backpropagation.
Adam N. McCaughanBakhrom G. OripovNatesh GaneshSae Woo NamAndrew DienstfreySonia M. BuckleyPublished in: CoRR (2023)
Keyphrases
- back propagation
- online training
- neural network
- weight update
- artificial neural networks
- feed forward
- learning rules
- training algorithm
- online learning
- hidden layer
- fuzzy logic
- backpropagation algorithm
- feedforward neural networks
- neural nets
- feed forward neural networks
- bp neural network
- bp algorithm
- multilayer perceptron
- backpropagation neural networks
- learning algorithm
- technical support
- activation function
- e learning
- cascade correlation
- connection weights
- error function
- multilayer perceptron neural network
- backpropagation neural network
- trained neural network
- levenberg marquardt
- error back propagation
- computer systems
- pattern recognition
- motor imagery
- human machine interaction
- control system
- learning rate
- real time
- machine learning
- artificial intelligence
- recurrent neural networks
- fault diagnosis
- mlp neural networks