On-chip training of recurrent neural networks with limited numerical precision.
Taesik NaJong Hwan KoJaeha KungSaibal MukhopadhyayPublished in: IJCNN (2017)
Keyphrases
- recurrent neural networks
- echo state networks
- recurrent networks
- feedforward neural networks
- neural network
- feed forward
- hidden layer
- feed forward neural networks
- reservoir computing
- high speed
- low cost
- artificial neural networks
- neural model
- complex valued
- long short term memory
- training process
- cascade correlation
- hebbian learning
- real time
- analog vlsi