Reducing Exposure Bias in Training Recurrent Neural Network Transducers.
Xiaodong CuiBrian KingsburyGeorge SaonDavid HawsZoltán TüskePublished in: Interspeech (2021)
Keyphrases
- recurrent neural networks
- recurrent networks
- feedforward neural networks
- echo state networks
- neural network
- feed forward
- feed forward neural networks
- hidden layer
- reservoir computing
- neural network structure
- long short term memory
- artificial neural networks
- complex valued
- nonlinear dynamic systems
- training process
- training phase
- training set
- training algorithm
- neural model
- biologically inspired
- test set
- training examples
- training samples
- adaptive neural
- hebbian learning
- learning algorithm
- real time