Improving the Accuracy of Deep Neural Networks Through Developing New Activation Functions.
Marina Adriana MercioniAngel Marcel TatStefan HolbanPublished in: ICCP (2020)
Keyphrases
- activation function
- neural network
- back propagation
- feed forward
- artificial neural networks
- feed forward neural networks
- neural nets
- neural architecture
- hidden layer
- hidden neurons
- multilayer perceptron
- learning rate
- hidden nodes
- radial basis function
- network architecture
- pattern recognition
- connection weights
- input space
- genetic algorithm
- recurrent neural networks
- basis functions
- training phase
- multi layer perceptron
- feedforward neural networks
- artificial intelligence
- dynamic programming