Logical Activation Functions: Logit-space equivalents of Boolean Operators.
Scott C. LoweRobert EarleJason d'EonThomas TrappenbergSageev OorePublished in: CoRR (2021)
Keyphrases
- activation function
- boolean operators
- neural network
- hidden layer
- artificial neural networks
- back propagation
- feed forward
- network architecture
- neural architecture
- neural nets
- extended boolean
- linear combination
- radial basis function
- dimensionality reduction
- multilayer perceptron
- input space
- high dimensional
- reinforcement learning
- information retrieval