Optimization and Generalization of Shallow Neural Networks with Quadratic Activation Functions.
Stefano Sarao MannelliEric Vanden-EijndenLenka ZdeborováPublished in: NeurIPS (2020)
Keyphrases
- activation function
- neural network
- feed forward
- back propagation
- artificial neural networks
- hidden layer
- feed forward neural networks
- neural nets
- hidden neurons
- connection weights
- multilayer perceptron
- learning rate
- hidden nodes
- feedforward neural networks
- neural architecture
- network architecture
- multi layer perceptron
- optimization algorithm
- global exponential stability
- pattern recognition
- fuzzy logic
- recurrent neural networks
- global optimization
- radial basis function
- optimization method
- reinforcement learning
- fuzzy sets
- objective function
- basis functions
- fault diagnosis
- learning algorithm
- fuzzy neural network
- support vector machine
- hardware implementation