AutoLrOpt: An Efficient Optimizer Using Automatic Setting of Learning Rate for Deep Neural Networks.
Mohamed MerrouchiKhalid AtifiMustapha SkittouYoussef Ben YoussefTaoufiq GadiPublished in: IEEE Access (2024)
Keyphrases
- learning rate
- neural network
- hidden layer
- activation function
- convergence rate
- training algorithm
- learning algorithm
- error function
- convergence speed
- multilayer neural networks
- adaptive learning rate
- rapid convergence
- fuzzy neural network
- feed forward
- uniform convergence
- back propagation
- feedforward neural networks
- fuzzy logic
- neural network model
- artificial neural networks
- feed forward neural networks
- query optimization
- global optimization
- genetic programming
- optimization algorithm
- training process
- fault diagnosis
- recurrent neural networks
- training speed
- multilayer perceptron