A Fast Approximation of the Hyperbolic Tangent When Using Posit Numbers and Its Application to Deep Neural Networks.
Marco CococcioniFederico RossiEmanuele RuffaldiSergio SaponaraPublished in: ApplePies (2019)
Keyphrases
- neural network
- artificial neural networks
- pattern recognition
- back propagation
- error bounds
- approximation methods
- self organizing maps
- closed form
- neural network model
- search algorithm
- approximation algorithms
- recurrent neural networks
- neural nets
- multi layer
- training algorithm
- learning rules
- feedforward neural networks
- competitive learning
- deep learning
- approximation error
- continuous functions
- fuzzy logic
- tangent distance