Automated Design of Linear Bounding Functions for Sigmoidal Nonlinearities in Neural Networks.
Matthias KönigXiyue ZhangHolger H. HoosMarta KwiatkowskaJan N. van RijnPublished in: CoRR (2024)
Keyphrases
- automated design
- neural network
- linear functions
- nonlinear functions
- upper bound
- activation function
- engineering design
- evolutionary design
- security protocols
- rbf network
- artificial neural networks
- feed forward neural networks
- fuzzy logic
- basis functions
- lower bound
- back propagation
- genetic algorithm
- mechanism design
- vc dimension
- feed forward
- real time