T-LRA: Trend-Based Learning Rate Annealing for Deep Neural Networks.
Samira PouyanfarShu-Ching ChenPublished in: BigMM (2017)
Keyphrases
- learning rate
- neural network
- activation function
- hidden layer
- backpropagation algorithm
- training algorithm
- convergence rate
- adaptive learning rate
- learning algorithm
- error function
- rapid convergence
- weight update
- fuzzy neural network
- feed forward neural networks
- simulated annealing
- convergence speed
- genetic algorithm
- multilayer neural networks
- feedforward neural networks
- artificial neural networks
- multi layer perceptron
- weight vector
- back propagation
- recurrent neural networks
- feed forward
- neural network model
- genetic programming
- fuzzy logic
- state space
- evolutionary algorithm