Positional Encoding Helps Recurrent Neural Networks Handle a Large Vocabulary.
Takashi MoritaPublished in: CoRR (2024)
Keyphrases
- recurrent neural networks
- feed forward
- neural network
- echo state networks
- recurrent networks
- complex valued
- reservoir computing
- cascade correlation
- continuous speech recognition
- artificial neural networks
- neural model
- feedforward neural networks
- hidden layer
- speech recognition
- nonlinear dynamic systems
- long short term memory
- encoding scheme
- hebbian learning
- positional information
- input output